Synthesis and Photophysical Characterization of an Artificial Photosynthetic Reaction Center

Exhibiting Acid-Responsive Regulation of Charge Separation

by

Ian Pahk

A Dissertation Presented in Partial Fulfillment of the Requirements for the Degree

Doctor of Philosophy

Approved April 2015 by the Graduate Supervisory Committee:

Devens Gust, Chair

Ian Gould Vladimiro Mujica

ARIZONA STATE UNIVERSITY

May 2015

i

ABSTRACT

Non-photochemical quenching (NPQ) is a photoprotective regulatory mechanism

essential to the robustness of the photosynthetic apparatus of green plants. Energy flow within

the low-light adapted reaction centers is dynamically optimized to match the continuously

fluctuating light conditions found in nature. Activated by compartmentalized decreases in pH

resulting from photosynthetic activity during periods of elevated photon flux, NPQ induces rapid

thermal dissipation of excess excitation energy that would otherwise overwhelm the apparatus’s

ability to consume it. Consequently, the frequency of charge separation decreases and the

formation of potentially deleterious, high-energy intermediates slows, thereby reducing the threat

of photodamage by disallowing their accumulation. Herein is described the synthesis and

photophysical analysis of a molecular triad that mimics the effects of NPQ on charge separation

within the photosynthetic reaction centers. Steady-state absorption and emission, time-resolved

fluorescence, and transient absorption spectroscopies were used to demonstrate reversible

quenching of the first singlet excited state affecting the quantum yield of charge separation by

approximately one order of magnitude. As in the natural system, the populations of unquenched

and quenched states and, therefore, the overall yields of charge separation were found to be

dependent upon acid concentration.

ii ii

ACKNOWLEDGEMENTS

First and foremost, I would like to thank Professor Devens Gust for providing me with the

distinguished privilege of learning from his enormous wealth of knowledge in spectroscopy,

photochemistry, and photophysics and for giving me the freedom, both through funding and in

practice, to develop as a synthetic chemist who obsesses over purity, yields, and crystalline

products. I would also like to thank Professors Tom and Ana Moore, together with Devens, for

providing a diverse, multi-disciplinary learning environment benefiting from their own expertise, as

well as those of the combined (GMM) research group members. Thanks go to all of the GMM

researchers with whom I interacted over the course of my graduate studies: Ant, Jeffrey, Bobby,

Jesse, Matthieu, Manuel, Michael, John, Smitha, Kul, Ben, Chris, Dustin, Brad, Dalvin, Jaro,

Katie, Chelsea, Marely, Raquel, Nino, Max, Lucie, and Graeme. In particular, I would like to thank

Ant for her kind acceptance in sharing both an office and lab space with me, and my

ever-expanding messes. Also, I want to thank her for bearing first hand witness to the enormous

range of moods and emotions, both high and low, that I showed up to work with over the last five

years. On a brighter note, the countless fits of delirious laughter in the lab and office are some of

my best memories of graduate school that I will never remember the causes of. In similar regard,

I would also like to thank Jeffrey for dealing with me in close quarters for the majority of my

graduate career, and also, for providing a constant stream of daily entertainment in the lab.

Special thanks go to Yuichi Terazono, who laid virtually all of the groundwork from which

my dissertation work was based. Yuichi is one of the finest synthetic chemists I have had the

pleasure of working with and I am forever grateful for his unending willingness to provide me with

synthetic advice and encouragement, in addition to teaching me everything he had learned about

working with these molecules. Of particular note – he taught me how to synthesize, purify, and

manipulate the sometimes-impossibly-finicky rhodamine dye that is the crux of the novelty of this

work. On this note, I would like to acknowledge Kul Bhushan for discovering this dye, which suits

the photophysical demands of this project so very well.

iii iii

Special thanks go to Paul Liddell, who taught me so much about working with porphyrins

and polyaromatic compounds and especially palladium coupling reactions thereof. Paul is,

perhaps, the epitome of what a synthetic organic chemist should be, and I think that I will never

gain the appreciation and understanding of organic molecules that he gives the impression of

having had since the first day he stepped up to a bench. I would like to thank Paul for sharing his

prolific experience and giving me synthetic advice, talking strategies and techniques, occasionally

providing me with chemicals and distilled solvents, and for flame sealing pressure vessels so that

my precious compounds would not wind up splattered all over the inside of my hood.

Special thanks go to Gerdenis Kodis, who taught me much of what I know about

photophysics and who performed all of the time-resolved fluorescence and transient absorption

experiments and data analysis and provided the mechanistic interpretations described later in this

work. Gerdenis was always willing to reteach me the aspects of ultrafast non-linear spectroscopy

and photophysical analyses that I forgot, over and over and over. I would also like to thank

Gerdenis for always appeasing my requests for additional experiments and for walking me

through his reasoning until I was fully convinced by his interpretations of the data.

Finally, I would like to thank my friends and family for the unending love and support they

showed me during this period of my life. I am forever grateful that my mother and father have

always supported me in choosing my own path in life. I would also like to express my gratitude for

my sister, whom I don't see often, but who somehow seems to share all of my weirdest quirks. It

is nice to know there is at least one other person who is weird, almost just like me. I would

especially like to thank Josh, Eri, Io, Jeffe, Kayla, Brian, Jamie, Logan, Brian, and Jenna for

always being there when I needed you to listen, to laugh, to tell me when I'm wrong, to remind me

to just relax; I'm doing fine, to help me get away, and to help me find myself again and again (and

again). You all mean the world to me and I am so grateful to have you as companions on this

ride.

iv iv

This work was funded by the Helios Solar Energy Research Center, which is supported

by the Director, Office of Science, Office of Basic Energy Science of the U.S. Department of

Energy under Contract No. DE-AC02-05CH11231; the Office of Basic Energy Sciences, Division

of Chemical Sciences, Geosciences, and Energy Biosciences, Department of Energy under

contract DE-FG02-03ER15393; and the Center for Bio-Inspired Solar Fuel Production, an Energy

Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of

Basic Energy Sciences under Award Number DE-SC0001016.

v v

TABLE OF CONTENTS

Page

LIST OF TABLES ................................................................................................................... vi

LIST OF FIGURES ................................................................................................................. vii

LIST OF SCHEMES ............................................................................................................... viii

CHAPTER

1 NON-PHOTOCHEMICAL QUENCHING AND ARTIFICIAL MIMICS THEREOF ... 1

Overview ....................................................................................................... 1

Photosynthesis and the Need for Photoprotection ....................................... 2

Non-Photochemical Quenching .................................................................... 3

Mimicry of Non-Photochemical Quenching .................................................. 6

2 SYNTHESIS OF AN ACID-RESPONSIVE REACTION CENTER MODEL ............ 15

Retrosynthetic Analysis and Synthetic Strategies ........................................ 15

Synthetic Pathways ...................................................................................... 17

3 PHOTOPHYSICAL AND KINETICS ANALYSES ................................................... 30

Overview ....................................................................................................... 30

Theories of Energy and Electron Transfer ................................................... 30

Conceptual Descriptions of Spectroscopic Techniques ............................... 33

Photophysical Analysis and Discussion ....................................................... 35

Kinetics Analysis ........................................................................................... 49

Concluding Remarks .................................................................................... 55

REFERENCES ....................................................................................................................... 57

APPENDIX

A SYNTHETIC METHODS AND CHARACTERIZATION DATA ................................ 63

B SPECTROSCOPIC METHODS .............................................................................. 91

vi vi

LIST OF TABLES

Table Page

3.1 Time-Resolved Fluorescence Exponential Decay Fitting Results for 1 and 2 ........... 38

vii vii

LIST OF FIGURES

Figure Page

1.1 A Photochromically Controlled Antenna-Reaction Center Model .............................. 8

1.2 An Acid-Responsive Multi-Porphyrin Antenna Model ................................................ 10

1.3 An Acid-Responsive Reaction Center Model ............................................................ 13

1.4 Model Compounds for Photophysical Analysis ......................................................... 14

2.1 Proposed Products of a Porphyrin Condensation with Scrambling ........................... 20

3.1 A Schematic Marcus Theory Description of Photoinduced Electron Transfer ........... 31

3.2 Titration of 1 in Dichloromethane With Acetic Acid Monitored by Steady-State

Absorption and Emission Spectroscopies ................................................................. 36

3.3 Time-Resolved Fluorescence Decays for 1 and 2 With Varying Concentrations of

Acetic Acid in Dichloromethane ................................................................................. 38

3.4 Time-Resolved Fluorescence Decay for 1 in Toluene .............................................. 39

3.5 Transient Absorption Decay Associated Spectra Showing Formation of a Long-Lived

Charge Separated State for 1 in Dichloromethane ................................................... 42

3.6 Transient Absorption Decay Associated Spectra Showing Charge Recombination for

1 in Dichloromethane ................................................................................................ 43

3.7 Transient Absorption Decay Associated Spectra for 1 in Toluene ............................ 44

3.8 Transient Absorption Decay Associated Spectra Showing the Relaxation Mechanism

of the Open Dye for 3 in Dichloromethane ................................................................ 46

3.9 Schematic Representation of the Relaxation Pathway of the Open Dye .................. 47

3.10 Transient Absorption Decay Associated Spectra Showing Quenching of the Porphyrin

Singlet Excited States of 1 and 2 by the Open Dye in Dichloromethane .................. 48

3.11 Energy Level Diagram for 1 in Dichloromethane ....................................................... 51

viii viii

LIST OF SCHEMES

Scheme Page

2.1 Retrosynthetic Analysis of Triad 1 ............................................................................. 16

2.2 A Proposed Synthetic Pathway to 1 .......................................................................... 18

2.3 Functionalization of Aldehydes and Alternative Porphyrin Condensations ............... 22

2.4 Synthetic Approaches to a Non-Symmetrical Hexaphenylbenzene .......................... 24

2.5 The Final Synthetic Pathway Followed to Obtain 1 ................................................... 26

2.6 The Synthetic Pathway Followed to Obtain 2 ............................................................ 29

1 1

Chapter 1

Non-Photochemical Quenching and Artificial Mimics Thereof

Charge separation is the primary light-driven reaction in photosynthesis. Solar energy is

converted to electrochemical potential in the form of charge separated states within

photosynthetic reaction centers. This provides driving force for water oxidation; the critical step

toward phototrophic production of chemical fuels. The ability for the photosynthetic apparatus, a

membrane-bound nanomolecular factory, to regulate the efficiency of energy conversion in

response to solar energy flux is paramount to the survival of photosynthetic organisms. In

parallel, developing responsive, self-regulating systems is important to address in the design of

practical and robust nanomolecular devices.1 In this regard, solar energy technologies based on

organic molecular components are expected to require regulatory and photoprotective

mechanisms in order to maximize their operating efficiencies and lifespans.2, 3 Biomimicry is

recognized as an important approach to the design of complex nanomolecular systems.4, 5

Nature's answer to photoprotection and the regulation of photosynthetic activity at the molecular

level is a mechanism known as non-photochemical quenching (NPQ), which has been the subject

of rigorous investigations for over 40 years. While many of the quantifiable operating parameters

of NPQ have been well characterized, our mechanistic understanding declines with- and is limited

by- our ability to resolve and deconvolute complex spectroscopic information to a molecular level.

This is a major obstacle in studying any natural system, as many mechanistic details change or

simply vanish when moving from highly complex holistic samples to relatively simplified in vitro

preparations.6 Approaching this problem from the bottom up, the development of small-molecule

functional analogs and, in a stepwise fashion building up a biomimetic environment around them,

may help to advance our understanding of how individual inter- and intramolecular interactions

influence the overall functioning of complex natural systems. Toward these ends, the research

described in this dissertation involves the synthesis and photophysical characterization of a

molecular triad that exhibits the responsive, self-regulation of charge separation that is a hallmark

2 2

of NPQ. A brief overview of photosynthesis and the necessity for photoprotection, the state of

knowledge regarding NPQ, and the most relevant examples of its mimicry in model systems

follows.

Photosynthesis is the biochemical process by which phototrophic organisms use solar

energy to pay the thermodynamic cost of sustained life. A series of photon capture, energy

transfer, charge separation, and energy conversion reactions ultimately produce the chemical

equivalents needed to maintain non-equilibrium, reduced (living) states in an oxidizing

environment. Chlorophyll, a porphyrin, serves as a primary photosynthetic pigment that fills dual

roles as light absorbers and as electron donors. Photo-excitation of the P680 special chlorophyll

pair within Photosystem II (PSII) generates its first excited singlet state, 1P680, which then readily

undergoes one of the most significant charge separation reactions in photosynthesis; donation of

an electron, via a series of electron transfer reactions, to a quinone (Q) thus generating P680!+, a

powerful oxidizer. The oxidizing potential of P680!+ is transferred to the metal center of the

oxygen-evolving complex (OEC) where it is used to split water by the following chemical

equation:

2 H2O + 4 hν → O2 + 4 H+ + 4 e-

Protons liberated by this reaction are selectively released into the lumen side of the

photosynthetic membrane, thereby generating an electrochemical potential gradient. Reduced

quinones, Q!-, shuttle electrons unidirectionally to Photosystem I (PSI) in a series of electron

transfer steps across multiple intermediaries. The exchange of electrons between each species

also drives protons across the photosynthetic membrane and into the lumen, further building the

potential gradient. The electrons that PSI receives from this transport chain are reenergized by a

light-based reaction so that they can be used in NADPH production. Simultaneously, ATP is

produced by channeling protons back across the photosynthetic membrane thereby consuming

3 3

the potential stored in the gradient. The energy stored in both NADPH and ATP is utilized in the

reduction of CO2 to form carbohydrates, the final step in photosynthetic energy conversion.

Light harvesting antennae, namely light harvesting complex II (LHCII), are employed in

the photosynthetic apparatus of plants to increase photon capture and funnel excitation energy

into PSII. This has the net effect of increasing the frequency of early charge separation reactions

and water oxidation in PSII. In this way, photosynthesis has evolved to be adaptive to intermittent

periods of low light intensity. Conversely, periods of high-intensity irradiation lead to higher

frequencies of photon capture and charge separation to the point of overwhelming the maximum

turn over frequencies of slower downstream electron transfer reactions and generation of NADPH

and ATP.7 Consequently, the high-energy intermediates involved in these reactions accumulate

and, if not efficiently consumed, can cause irreversible damage to- and permanent deactivation

of- PSII.8-10 Photo-damage to PSII is primarily associated with P680.9 If the rate at which 1P680 is

oxidized far exceeds that at which electrons are donated back to P680!+ from the OEC, the

resulting increase in the lifetime of P680!+ allows it to oxidize pigment molecules and amino acid

residues within its immediate vicinity.8, 9 Alternatively, if the reduced quinone pool is too large,

signifying the inability for downstream reactions to keep pace with the light based reactions within

PSII, the P680!+-Q!- charge separated state can recombine leading to formation of the triplet,

3P680.8, 9 This triplet can interact with oxygen to form singlet oxygen, 1O2, which will react with

many components of the photosynthetic apparatus leading to pigment bleaching and, eventually,

PSII deactivation.8, 9

Given the nature of interactions between organic pigments, light, and oxygen, the

evolutionary emergence of oxygenic phototrophs stipulated a compulsory requirement for layering

of multi-level regulatory and protective mechanisms to prevent the primary source of energy, the

sun, from engendering their destruction. Non-photochemical quenching (NPQ) is a

photoprotective regulatory mechanism that modulates the efficiency of energy transfer from LHCII

to PSII in response to fluctuating light intensity. During periods of low photon flux, it is necessary

for solar energy conversion to be performed with maximal efficiency in order to provide the

4 4

organism with enough energy and reducing capacity to maintain homeostasis. Therefore, the

photosynthetic apparatus has adapted to fulfill this demand by dividing electron transport into a

series of slower but highly energy efficient steps. Conversely, during periods of high photon flux,

the initial photon capture and energy transfer steps of photosynthesis make available an excess

of excitation energy to the reaction centers that downstream electron transport and chemical fuel

producing reactions have no innate way of processing. As the release of protons into the

thylakoid lumen continues to outpace the ability for ATP synthesis to utilize the potential stored in

the proton gradient, the resulting decrease in lumen pH signals an over abundance of solar

energy and proportional activation of NPQ.6, 7, 11 Time-resolved fluorescence spectroscopy has

shown that NPQ attenuates the excited state lifetime of LHCII chlorophylls, thereby reducing the

quantum efficiency of energy transfer to PSII.12-14 Consequently the frequency of charge

separation in the reaction center also decreases (as does the charge separation efficiency

relative to total photons captured by the apparatus), thereby bringing light-to-potential energy

conversion into equilibrium with potential-to-chemical energy conversion.15, 16 Exposure to low

light conditions allows for consumption of the proton gradient, which is reciprocated by

proportional deactivation of NPQ and return to elevated efficiencies of charge separation. In this

way NPQ is dynamically responsive to fluctuations in light intensity.

NPQ activates one or more decay pathway(s) by which excess excitation energy

collected by the antennae can be harmlessly dissipated as heat rather than being used to

generate charge separated states.15 While decreases in lumen pH have long been known to play

a causative role in the activation of NPQ, the molecular nature of this responsiveness remains

uncertain.11, 17 As such, the complex and multi-layered feedback mechanisms that allow for highly

adaptive and robust control over NPQ are not fully understood.6 NPQ activation is thought to

induce reorganization and aggregation of protein complexes along the photosynthetic membrane,

for which several models have been proposed.18, 19 In relation to these spatial and superstructural

changes, factors including the pH-responsive cycling of xanthophyll carotenoids and protonation

5 5

of an auxiliary protein, PsbS, have been implicated as modulators of the rates of activation and

relaxation of quenching as well as the pH threshold at which NPQ activation is initiated.20-22

The identities of the photophysical mechanism and molecular quencher responsible for

the rapid thermal dissipation observed in NPQ have been subjects of debate for over 20 years.6

Discernment of potential quenching mechanisms is restricted by the kinetic parameters set by the

experimentally observed fluorescence lifetimes of LHCII chlorophyll; around 2 ns in the

unquenched state and roughly 0.5 ns in the quenched state.12 In the absense of

interchromophore interactions, the first singlet excited state of chlorophyll is known to decay by

fluorescence, internal conversion, and intersystem crossing; processes more rigorously defined in

Chapter 3. The decay lifetimes associated with intersystem crossing are generally much longer

than the observed quenched lifetime making involvement of this type of process unlikely. On the

other hand changes in the local protein environment of a particular chlorophyll could conceivably

transform it into a quenching species by increasing its rate of decay by internal conversion,

thereby fulfilling this kinetic requirement.6 Structural changes to the local environment of a

chlorophyll could also bring another highly quenched pigment into sufficiently close proximity to

allow for energy transfer and dissipation.6 Ultrafast spectroscopic studies focusing on carotenoid

and chlorophyll pigments in vivo and in vitro have yielded several prominent but disputed theories

regarding the molecular identity of the quencher. Earlier time-resolved fluorescence work

suggested formation of a chlorophyll-chlorophyll dimer within LHCII complexes as the primary

NPQ quencher based on the similarly red-shifted emission characteristics of LHCII and those of

known chlorophyll aggregates with short excited state lifetimes.12, 19, 23 More recent transient

absorption and computational studies assign the role of quencher to a xanthophyll carotenoid.

Some implicate coherently coupled chromophores and quenchers over which an excited state

wavefunction is distributed.24, 25 Such coherent excited states are proposed to decay by charge

transfer in which rapid charge separation and subsequent recombination yield the ground state

with the release of heat.26, 27 Others have suggested an incoherent energy transfer mechansim

where the excited state discretely "hops" from one pigment to another until reaching a quencher

6 6

with an intrinsically short excited state lifetime.28 One of the most recent reports revisited

chlorophyll-chlorophyll charge transfer complexes (rather than chlorophyll-carotenoid) under a

coherently coupled regime to explain the chromophore coupling and charge transfer

characteristics that had been previously reported.29 While each of these theories is supported by

experimental and in some cases computational results, they are all also opposed by contradictory

interpretations and lack the support of undisputable substantiating evidence in their favor. In

short, the physical environment that surrounds the quenching process in NPQ is intricately

complex, which makes detailed and biologically relevant studies difficult to perform. Alternatively,

research has turned toward developing simplified model systems with the goal of adapting them

to increasingly complex biomimetic local environments in order to identify specific operating

parameters and interaction-based photophysical phenomenon that activate different quenching

processes in a highly controllable manner.

In the last decade, the laboratories of Professors Devens Gust, Ana L. Moore, and

Thomas A. Moore have reported several small molecules that model various aspects of NPQ.

The earliest work in this area involved quenching of chromophore excited states through

interactions with synthetic carotenoids.30, 31 Prior to the work described in this dissertation, there

were only two published examples of small molecules that demonstrate self-regulating quenching

behavior. The first was an antenna-reaction center model that utilized a photonic switch to

provide non-linear control over its charge separation efficiency in an inverse relationship with

incident light intensity, a good analog to the overall behavior observed in NPQ.32 In an effort to

more closely mimic the response mechanisms of NPQ, an antenna model was designed around

an acid-sensitive quenching switch that could shorten the excited state lifetime of the antenna

chromophores upon its protonation at elevated acid concentrations.33 What follows are reviews of

these reports to provide context for the subject of this dissertation. The spectroscopic analyses of

these systems rely heavily on theoretical approaches to charge separation in molecular donor-

acceptor manifolds put forth by Rudolph A. Marcus, and through-space singlet-singlet energy

transfer put forth by Theodor Förster.34-36 Conceptual summaries of these theories as well as

7 7

theoretical descriptions of the optical spectroscopic techniques used in this research are provided

in Chapter 3.

The first model system to demonstrate adaptive regulation over charge separation,

shown in Figure 1.1, utilized a photochromic dihyroindolizine (DHI)/betaine (BT) switch, originally

developed for molecular logic applications, to modulate the quantum yield of charge separated

states formed relative to incident light intensity.32, 37 Designed around established operating

parameters for the photochrome, the reaction center was comprised of two

bis(phenylethynyl)anthracene (BPEA) antenna moieties capable of singlet-singlet energy transfer

to a porphyrin electron donor in essentially unity yield, thereby expanding the spectral cross

section of light that could lead to formation of a charge separated state via electron transfer to the

fullerene acceptor. The more stable spirocyclic DHI form of the switch does not interact with

either the BPEA or porphyrin moieties, and is therefore photophysically passive in this capacity.

DHI does, however, absorb blue light, which induces photoisomerization to its BT form, a good

energy acceptor for the first singlet excited state of the porphyrin. Rapid energy transfer to BT

thus effectively down-regulates the capacity for formation of the porphyrin-cation-fullerene-anion

charge separated state, P!+-C60!−. As the closed form is favored thermodynamically, it can be

restored by thermal isomerization at ambient temperatures in the dark with a 37 s time constant.

Isomerization could also be effected by irradiation with red light, albeit inefficiently. In this way,

the photochromic switch is able to non-linearly transduce incident white light intensity as charge

separation efficiency.

In the closed form, the reaction center exhibits charge separation by electron transfer to

the fullerene with a 2 ns time constant. This gives a ‘low light’ charge separation efficiency of 82%

with respect to the porphyrin excited state. Energy transfer to the open, BT form of the

photochrome was found to have a time constant of 33 ps thereby reducing the charge separation

efficiency to 1%, thus demonstrating its efficacy as a quenching switch. Simulations of its

behavior under real operating conditions were generated first by monitoring absorption from the

charge separated state over 15 cycles of light and dark periods to demonstrate reversibility and

8 8

HN

O

NH

N

O

NNC

CN

O

NNC

CN

Blue hν Red hνor ∆

Closed(DHI)

Open(BT)

NNHN HN

OMe

MeO

OMeMeO

Figure 1.1 A photochromically controlled antenna-reaction center model. With the closed (DHI) form of the switch, excitation of the BPEA or porphyrin moieties readily leads to formation of the P!+-C60

!− charge separated state. Photoisomerization to the open (BT) form leads to quenching of the porphyrin first singlet excited state via rapid energy transfer to BT.

9 9

robustness, and second by determination of charge separation efficiencies at varying intensities

of white light to demonstrate its dynamic responsiveness. In the latter experiment the lowest

charge separation efficiency of the total population of closed and open isomers was 37%, less

than half of the maximum efficiency observed under 'low light' conditions. While this model

system achieves its intended functionality, it is only responsive to the blue wavelengths of light

absorbed by DHI rather than total light absorbed by the antenna-reaction center complex. The

mechanism by which quenching is activated is not biomimetic in the sense that it does not rely on

the relay of information regarding system performance via proton activity. This limitation of

responsiveness to light absorbed specifically by the photochrome eliminates the capacity for self-

regulation in response to a hypothetical downstream product of reaction center activity.

A multi-porphyrin antenna linked to a rhodamine dye, shown in Figure 1.2, is the first

reported model system to exhibit acid-responsive regulation of its excited state lifetime in an

analogous fashion to NPQ.33 Development of this antenna stemmed from the discovery of a novel

rhodamine dye that possesses photophysical properties in its protonated, open form that make it

an excellent energy acceptor for the singlet excited states of zinc tetra-aryl porphyrins. This acid-

responsive colorimetric behavior is a hallmark of rhodamine dyes making them useful as pH

indicators. Titration of a solution of the model antenna with acetic acid was monitored by

steady-state absorption and emission spectroscopies to observe the shift in equilibrium from the

closed (colorless) form to the open (blue) form of the dye. The amplitude of a new absorption

band at 656 nm, diagnostic for the open dye, was found to be directly influenced by the

concentration of acid. The impact of this equilibrium shift was observed in the corresponding

emission spectra, in which the porphyrin fluorescence intensity was reduced in proportion to the

population of antennae that had been converted to the open dye form. Thus, it was demonstrated

that this dye effectively models the pH-responsive excitation energy quenching that characterizes

NPQ.

Photophysical analysis of the antenna provides a quantitative description of the efficiency

of the dye as an excitation energy quencher. In its closed form, the dye imparts no effect on the

10 10

N

N N

NZn

N

OMeO

O

O NN

OHO

O

N

N

NNN

NZn

NN

NN ZnN

NN

NZn

NNN

NZn

+[H+] -[H+]

Closed

Open

Figure 1.2 An acid-responsive multi-porphyrin antenna model. In the closed form, the dye imparts no effect on photophysical behavior of the antenna. The open, protonated dye acts as an energy sink for the first singlet excited states of the zinc porphyrins, decreasing their fluorescence lifetimes via rapid singlet-singlet energy transfer.

11 11

excited state lifetime of zinc porphyrins. Thus, in a neutral solution the antenna was found to

possess a fluorescence decay lifetime of 2.1 ns, characteristic of a typical zinc porphyrin.

Treatment with an excess of acetic acid resulted in new decay components with time constants of

10 ps and 39 ps. The Förster model for singlet-singlet energy transfer was used to calculate

theoretical rates of energy transfer from the zinc porphyrins to the dye allowing for assignment of

the 10 ps decay to energy transfer from the ortho-porphyrin and the 39 ps decay to energy

transfer from the para-porphyrin. Exponential decay fitting could not separate a component for the

meta-porphyrin, and so its decay is assumed to be mixed with those of the ortho- and para-

moieties. Aside from ascribing the mechanism of excited state quenching to Förster-type energy

transfer, the mechanism of energy dissipation by the dye was not investigated in detail. In

contrast to more common rhodamine dyes (such as rhodamine 6G) that are used as fluorescence

standards, no emission coud be detected from this novel dye by either steady-state or ultrafast

fluorescence measurements. Transient absorption spectroscopy of the isolated dye gave a 5 ps

excited state decay lifetime for its open form, which physically rationalizes the lack of detectable

emission. In summary, this work showed the efficacy of the novel rhodamine dye as an energy

sink for the excited states of zinc porphyrins. The kinetics analysis shows attenuation of the

antenna’s excited state lifetime by roughly two orders of magnitude, which suggests that the dye

could also effectively inhibit electron transfer to an electron acceptor given that the necessary

kinetic parameters are met.

Herein is described the synthesis and photophysical characterization of triad 1, shown in

Figure 1.3, a reaction center model comprised of a zinc porphyrin donor, a fullerene acceptor,

and the aforementioned rhodamine dye quenching switch. In a neutral solution, excitation of the

porphyrin leads to formation of a long-lived charge separated state via transfer of an electron to

the fullerene. As in the reported antenna compound the protonated, open form of the dye rapidly

quenches the porphyrin singlet excited state via singlet-singlet energy transfer. Consequently, the

quantum yield of charge separation decreases dramatically. Steady-state absorption and

emission, time-resolved fluorescence, and transient absorption spectroscopies were used to

12 12

evaluate the efficacy of energy transfer to the dye as a regulator of charge separation in addition

to providing a more complete description of the energy quenching mechanism. Model compounds

2 and 3, shown in Figure 1.4, were also prepared and spectroscopically characterized in parallel

with 1 in order to isolate the decay kinetics of photophysical processes associated with the

porphyrin, rhodamine, and fullerene in exclusion of one or both of the others.

13 13

NN N

N

O

HN

N O

Zn

R

N

O

O

O

NN

HO

O

O

NN

+ [H+]

1c 1o

e-

En

- [H+]R =

Figure 1.3 An acid-responsive reaction center model, triad 1, that can be reversibly interconverted between its closed (1c) to its open (1o) forms by protonation/deprotonation of the rhodamine dye (DC, DO). Excitation of the porphyrin moiety of 1c leads to transfer of an electron (e-) to the fullerene to form a long-lived charge separated state, DC-PZn

!+-C60!−.

Conversion to 1o allows for rapid singlet-singlet energy (En) transfer from 1PZn to DO thereby reducing the quantum yield of charge separation.

14 14

N NNN HN

O

N

Zn

O

O

ON

N

O

2

3N

OO

ON

N

O

Figure 1.4 Model compounds dyad 2 and dye 3 were prepared to aid in the photophysical analysis of triad 1. In acidic solutions, they also exist in open forms analogous to that of 1o.

15 15

Chapter 2

Synthesis of an Acid-Responsive Reaction Center Model

Synthesis of triad 1 constituted the majority of time and funding put toward this body of

work. This chapter details the most important efforts toward developing a reproducible synthetic

pathway that yielded a sufficient quantity of 1 in at accetably high purity for use in photophysical

investigations. The total synthesis of 1 involved twenty-four non-linear synthetic steps that were

all carried out independently by the author. However, twelve of these reactions had been

previously reported and are simply referenced in this dissertation. While most of the steps along

unsucessful synthetic pathways discussed herein were novel, several previously reported

reactions are also described as they are of strategic significance or were the sources of major

synthetic challenges that warrant extended discussion. Experimental methods for all reasonably

successful, novel reactions and any characteristic chemical identification data (1H NMR,

MALDI-TOF-MS, UV-Vis) that were obtained are provided in Appendix A.

Retrosynthetic analysis of 1 (Scheme 2.1) indicates the necessity to form an intermediate

porphyrin-hexaphenylbenzene compound (iii) bearing both secondary amine and carboxylic acid

functionalities for coupling the rhodamine and fullerene moieties, respectively. It was expected

that these coupling reactions needed to be performed in this order to ensure that the reaction

conditions would be compatible with the reactivities of all substituents present. The reaction used

to attach the rhodamine dye, developed by Yuichi Terazono, involves the use of a palladium

catalyst that could, potentially, react with the fullerene moiety and therefore needed to be carried

out before the fullerene was introduced.33 Solubility is a concern with carboxylic acid bearing

porphyrins in non-polar organic solvents such as toluene, as is required for this reaction. Thus, it

was assumed that an ester would need to be present (iii, R = alkyl) to overcome possible

solubility issues that could have prevented the coupling reaction leading to iv from occurring.

While benzoic acid-bearing porphyrins are usually condensed as their corresponding methyl

esters (ii), the hydrolytic conditions required to obtain the acid involve either a large excess of

16 16

N NNN HN

O

N

Zn

O

O

ON

N

O

N

1

N NNN

H2N

O

N O

O

ON

N

O

N

OH

N NNN

O

HN

M

O

O

ON

N

O

O

Br

N N

NN O

OX

O

R

iviii

ii i

M

MN N

NN O

O

X

M

Scheme 2.1 Retrosynthetic analysis of triad 1.

17 17

potassium hydroxide or heating in a mixture of hydrochloric and trifluoroacetic acids. Rhodamine

dyes are known to decompose hydrolytically in the presence of hydroxide bases and so the

former conditions would not be a viable option for the removal of this protecting group.38 There

was also concern that the indoline groups may be oxidized to more stable indoles with prolonged

exposure to elevated temperatures in the presence of a strong acid, thereby eliminating the

option of using the latter conditions. Therefore, an electron-rich 2,4,6-trimethylbenzyl ester

(Scheme 2.2) was selected because it could be removed in the presence of a relatively mild acid,

trifluoroacetic acid, at ambient temperature (Terazono, et al. unpublished). The most attractive

and elegant method to obtain the porphyrin-hexphenylbenzene intermediate ii involves formation

of the hexaphenylbenzene moiety via a thermally driven [2+4] Diels-Alder cycloaddition between

a diphenylacetylene-bearing porphyrin and tetraphenylcyclopentadienone (i) followed by

elimination of carbon monoxide. This methodology was the most obvious starting place given its

reported efficacy toward obtaining intermediate ii and several structural analogs bearing unique

substituents at the 2’ phenyl ring relative to that connected to the porphyrin.32, 33, 39

With these constraints in mind, the synthetic pathway shown in Scheme 2.2 was

proposed. Porphyrin 5, bearing a diphenylacetylene moiety at the 20-meso position, was

prepared by borontrifluoride diethyletherate catalyzed [2+2] condensation from mesityl

dipyrromethane and the corresponding aldehydes.39 The hexaphenylbenzene moiety was then

formed by way of the Diels-Alder cycloaddition mentioned above to obtain 6.39 Treatment with

zinc acetate afforded 7, which was subsequently coupled with p-anisidine via a Buchwald-Hartwig

palladium catalyzed aryl-amination reaction yielding the secondary amine 8. Hydrolysis with

potassium hydroxide gave the carboxylic acid, 9, which was then treated with a carbodiimide

coupling reagent, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDCI), in the

presence of 4-dimethylaminopyridine (DMAP) and 2,4,6-trimethylbenzyl alcohol to afford the

desired ester, 10. Following established protocols for the rhodamine dye, 11, the protected dyad

12 was obtained.33 Deprotection of the ester with trifluoroacetic acid was carried out, however an

inseparable mixture of products was formed.

18 18

BrO

H

NHHN

NH N

HNN

O

OH

O

O

OBr

N N

NN O

O

Br

M

O

N N

NN OR

O

NH

Zn

O

NH2

O

6 M = H2

7 M = Zn

8 R = Me

5

9 R = H

N N

NN O

O

NH

Zn

O

HO

N NNN O

O

N

Zn

O

O

ON

N

O

10

12

10

NH NHNN OH

O

N O

O

ON

N

O

13

H2NN

14

O

ON

N

OBr

11

N NNN HN

O

N

M

O

O

ON

N

O

N

1 M = Zn

15 M = H2

+

i

ii

iii

iv

v

vi

vii

viii

ix

x

4

Scheme 2.2 A proposed synthetic pathway to 1: i a) BF3OEt2/EtOH, CHCl3, RT b) DDQ, RT; ii) Ph2O, reflux; iii) Zn(OAc)2·2H2O/MeOH, CHCl3, RT; iv) Pd(OAc)2, Cs2CO3, P(tBu)3, toluene, reflux; v) KOH (aq), MeOH/THF, RT; vi) EDCI, DMAP, CH2Cl2, RT; vii) Pd(OAc)2, Cs2CO3, P(tBu)3, toluene, reflux; viii) TFA/CH2Cl2 ix) EDCI, DMAP, CH2Cl2; x) Zn(OAc)2·2H2O/MeOH, CHCl3.

19 19

The stabilized benzylic carbocation intermediate produced in this reaction has a long

lifetime that allowed it to react with the desired product, 13, to form single and double adducts of

the 2,4,6-trimethylbenzyl group (133 amu). These adducts were detected by MALDI TOF-MS

which showed that, in addition to the expected molecular ion peak at m/z = 1853, peaks at

m/z = 1985 and 2118 were present in relatively high intensities. These compounds ran as a

single spot by thin layer chromatography in all solvent systems that were screened. Resolution of

the mixture was attempted numerous times by flash chromatography and preparative thin layer

chromatography to no avail. A major issue with chromatographing compounds containing this

rhodamine dye is that, in the presence of the silica gel stationary phase, the dye is in equilibrium

between its closed (colorless) and open (blue) forms. This has the plainly visible effect of severe

band broadening during column chromatography. Other stationary phases including basic and

neutral alumina were found to be similarly ineffective. The newly formed carboxylic acid of 13 was

also suspected of hindering chromatographic resolution, and so treatment with EDCI and DMAP

in the presence of anilinofullerene 14 was carried out in hopes of fascilitating separation of the

desired product from this mixture. Unfortunately, the desired product still could not be separated

and the batch of material was abandoned. Further literature research on this type of deprotection

reaction revealed that cation scavengers such as triethylsilane and 1,3,5-trimethoxybenzene can

be added in excess to prevent this type of side reaction from occurring.32

Repetition of the porphyrin condensation yielded the expected statistical mixture of

trans-A2BC (5), trans-A2B2, and trans-A2C2 porphyrins shown in Figure 2.1. Following

chromatographic separation, the mass spectrum of the band containing 5 showed an additional

peak at m/z = 799 which was attributed to the A3B porphyrin shown in Figure 2.1. Repeated

chromatographic purification yielded a fraction free of this impurity. However, the 1H NMR

spectrum still indicated the presence of a porphyrin impurity based on observation of minor

signals that apparently arose from a distinct mesityl group. At this point the mass spectrum only

contained the expected molecular ion peak at m/z = 936, so it is assumed that the cis-A2BC

geometric isomer of 5 (Figure 2.1) may have been formed. This and the A3B porphyrin are

20 20

NH NHN

N

OO

Br

NH NHN

N

OO

NH NHN

N

Br

trans-A2BC

A3B cis-A2BC

NH NHN

N

OO

trans-A2B2

OO

NH NHN

N

Br

trans-A2C2Br

OO

Figure 2.1 Proposed products of the porphyrin condensation shown in Scheme 2.2 (reaction i) including the expected trans-A2BC (5), trans-A2B2, and trans-A2C2 porphyrins as well as the A3B and cis-A2BC porphyrins that could result from scrambling.

21 21

potential products of scrambling side-reactions, which are discussed in greater detail below. As

no chromatographic or recrystallization technique was found that could resolve this mixture, effort

was put into achieving resolution of the material via synthetic modifications. Conversion from the

free base to the zinc-coordinating tetrapyrrole is often used to aid in resolution of otherwise

inseparable porphyrin mixtures. In this case, there was no change noted. The mixture was also

carried through several subsequent steps of the synthetic pathway outlined in Scheme 2.2

without success.

The group of Jonathan S. Lindsey has published extensively on porphyrin synthesis

methodologies with several reports dedicated to the mechanism of scrambling and methods of

reducing its occurrence.40, 41 Essentially, dipyrromethanes and porphyrin condensation

intermediates can undergo acid-catalyzed fragmentation to yield pyrrolic and azafulvene

compounds. These species are then able to recombine to form products of unwanted

compositions and geometries. Scrambling occurs in varying degrees depending upon the acid

catalyst, its concentration, and the structures and concentrations of the reactants. Reducing the

concentration of the acid catalyst has been shown to reduce scrambling.40 Additionally, catalysis

with trifluoroacetic acid in dichloromethane (CH2Cl2) was reported as a ‘no scrambling’ method for

porphyrin condensations.40 Reaction conditions were screened using varying concentrations of

different acid catalysts in combination with varying concentrations of reactants, also to no end.

Subsequently, the porphyrin condensation reactions shown in Scheme 2.3 were carried out. The

amine-functionalized aldehyde 17 was formed by way of a Sonogashira coupling reaction of

acetylene 16 with 4-iodobenzaldehyde. A portion of this material was converted to the

trifluoroacetamide 18 by treatment with trifluoroacetic anhydride. Again, all variations of the

porphyrin condensation reaction conditions oulined above resulted in inseparable scrambled

mixtures.

These failed attempts to produce isomerically pure batches of porphyrins 5, 19, and 20

necessitated the need for a new strategic approach to the synthesis of a workable

porphyrin-hexaphenylbenzene compound. Several structurally analogous antenna and reaction

22 22

NH N

HNN O

ON

O

N

O

H

O

NHHNO

OH

OR

R

17 R = H

18 R = OF3C

19 R = H

20 R = OF3C

ii

HN

O

H

OH I

i

16

iii

Scheme 2.3 Functionalization of aldehydes and alternative porphyrin condensations: i) PdCl2(PPh3)2, CuI, THF/Et3N, RT; ii) TFAA, pyridine, CH2Cl2 0 °C to RT; iii) a) BF3OEt2/EtOH, CHCl3, RT b) DDQ, RT or a)TFA/CH2Cl2, RT b) DDQ, RT.

23 23

center model compounds had previously been prepared following different synthetic

methodologies. In one report, a cobalt-templated [2+2+2] cycloaddition of three diphenylacetylene

compounds was used to produce a porphyrin-substituted hexaphenylbenzene.33 However, this

method would not be selective for non-symmetrical substitution at the desired 2’ position relative

to that of the porphyrin as in the structure of triad 1. The yield of the desired compound would

therefore be drastically reduced due to the statistical mixture of potentially inseparable geometric

isomers that would form. Another report used an aldehyde-substituted hexaphenylbenzene as a

reactant in a porphyrin condensation reaction.32 While this method would solve the issue of

geometric selectivity, the yield was unacceptably low for the purpose of this project.32 In contrast,

Suzuki coupling reactions between meso-bromoporphyrins and aromatic boronate esters have

been reported in excellent yields.42-44 This strategy was pursued to form the requisite

hexaphenylbenzene-porphyrin intermediate via a novel hexaphenylbenzene boronate ester.

The synthetic efforts toward obtaining the desired hexaphenylbenzene boronate ester 28

are outlined in Scheme 2.5. Initial attempts to synthesize this non-symmetrical disubstituted

hexaphenylbenzene involved substitution of dibromo-hexaphenylbenzene 21 with p-anisidine

using a Buchwald-Hartwig reaction. A complex mixture of products was obtained, and while it did

appear to contain the desired compound 22, the yield was unacceptably low. The presence of two

bromine atoms allowed for catalysis to continue beyond the first substitution to yield the

di-substituted and various cross-coupled products. Elimination of this issue required that the

hexaphenylbenzene be formed by a [2+4] cycloaddition involving a non-symmetrically substituted

diphenylacetylene. To this end, diphenylacetylene 23 was readily formed via a Sonogashira

palladium coupling of acetylene 16 and 4-bromo-iodobenzene. Due to the high temperature of the

forthcoming Diels-Alder cycloaddition reaction, 260 °C, and the propensity for diphenylamines to

undergo oxidative degradation, formation of a thermally and oxidatively resilient amide was

desired.45, 46 The use of a trifluoroacetamide was an attractive option as it could be hydrolyzed

using a hydroxide base at ambient temperature, conditions that would simultaneously hydrolyze a

methylbenzoate porphyrin substituent (as was successfully used for conversion of 8 to 9).

24 24

Br

Br

NH2

ONH

O

Br

N

O

BrF3C

O

N

O

Br

O

CF3

21 22

24

25

HN

O

H HN

O

BrBrI

N

O

Br

O

N

O

O

Br

N

O

O

B

BO

O

O

O

BO

O

26

27

28

16 23

O

i

ii

v

vi

vii

iii

iv

Scheme 2.4 Synthetic approaches to the non-symmetrical hexaphenylbenzene: i) Pd(OAc)2, Cs2CO3, P(tBu)3, toluene, reflux; ii) PdCl2(PPh3)2, CuI, THF/Et3N, RT; iii) TFAA, pyridine, CH2Cl2 0 °C to RT; iv) Ph2O, reflux; v) acetyl chloride, pyridine, CH2Cl2 0 °C to RT; vi) Ph2O, reflux; vii) Pd(dppf)Cl2•CH2Cl2, KOAc, 1,4-dioxane, 100 °C.

25 25

However, trifluoroacetamides are also known to only possess moderate stability toward thermal

decomposition above 150 °C.47 Treatment of 23 with trifluoroacetic anhydride afforded the

protected acetylene 24. Formation of 25 by cycloaddition with tetraphenylcyclopentadieneone

was accompanied by heavy, unresolvable decomposition as noted in the 1H NMR spectrum of the

product-containing fraction isolated by column chromatography. Clearly, a more thermally stable

protecting group was required. Thus, acetamide 26 was prepared by treatment of 23 with acetyl

chloride. The non-symmetrical hexaphenylbenzene 27 was then formed without decomposition by

the cycloaddtion reaction. Final purification of this intermediate by recrystallization from refluxing

toluene yielded a 1:1 complex of 27 with toluene, as noted by 1H NMR after forcefull attempts to

remove solvent residues under strong vacuum at elevated temperatures. Synthesis of the desired

boronate ester 28 was completed by way of a palladium catalyzed coupling reaction of 27 with

bis(pinacolato)diboron. During chromatographic purification, significant decomposition of the

boronate ester to the boronic acid was noted if the rate of elution was too low.

With boronate ester 28 and bromoporphyrin 30 (prepared by zincation of the free base

2948) in hand, the sequential assembly of triad 1 (Scheme 2.5) was begun. First, optimal

conditions for the critical Suzuki coupling reaction to yield the desired porphyrin-

hexaphenylbenzene 31 were investigated. Suzuki reactions are often carried out in coordinating

organic solvents and, in particular, tetrahydrofuran (THF) has been used very successfully with

porphyrin reactants.42-44, 49 Initial trials using THF demonstrated that 31 could, in fact, be

produced by this method. However, replacement of the meso-bromine atom with hydrogen was

by far the favored reaction under these conditions suggesting that the intermediate

palladium-porphyrin complex underwent protic elimination before it could react with the

hexaphenylbenzene boronate ester. Given the likliehood that steric hinderance may have

affected the reactivity of one or both compounds, it was hypothesized that a higher reaction

temperature could improve the kinetic favorability of the desired coupling reaction versus

elimination. Indeed, the same reaction, when allowed to proceed in refluxing toluene for three

days, gave 31 in yields of 65-74%. Global deprotection of the ester and acetamide groups was

26 26

N NNN O

O

N

Zn

OO

O

ON

N

O

O

HO

N N

NN O

O

N

Zn

O

O

N N

NNBr

O

OM

N

O

O

BO

O

9

2831

33

N N

NN OH

O

NH

Zn

O

NH NHNN OH

O

N O

O

ON

N

O

13

H2NN

14

N NNN HN

O

N

M

O

O

ON

N

O

N

1 M = Zn

15 M = H2

vii

viii

O

ON

N

OBr

11

N N

NN O

O

NH

Zn

O

O32

ii

iii

iv

v

vi

30 M = Zn

29 M = H2i

Scheme 2.5 The final synthetic pathway followed to obtain 1: i) Zn(OAc)2·2H2O/MeOH, CHCl3, RT; ii) Pd(PPh3)4, K3PO4, toluene, reflux; iii) KOH/MeOH, THF, 75 °C; iv) EDCI, DMAP, CH2Cl2, RT; v) Pd(OAc)2, Cs2CO3, P(tBu)3, toluene, reflux; vi) TFA/CH2Cl2, RT; vii) EDCI, DMAP, CH2Cl2, RT; viii) Zn(OAc)2·2H2O/MeOH, CH2Cl2, RT.

27 27

attempted under both basic and acidic conditions. Alkaline hydrolysis was carried out by addition

of saturated methanolic potassium hydroxide to a THF solution of 31 followed by reflux for 12

days. The mass spectrum showed complete conversion to the carboxylic and secondary amine

as indicated by a single peak with m/z = 1384 corresponding to the mass of 9. Daily analysis of

the reaction progress had shown complete hydrolysis of the methyl ester and minimal hydrolysis

of the acetamide within the first day, indicating that the acetamide is overwhelmingly stable to this

type of hydrolytic cleavage. In an attempt to achieve the desired reaction in less time, a sample of

31 was dissolved in a 2:1 mixture of concentrated hydrochloric and trifluoroacetic acids and held

at 80 °C in a sealed vessel. After several days, analysis of the reaction mixture indicated that

virtually no hydrolysis of the acetamide had occurred. Reiteration of the alkaline hydrolysis

procedures in sealed pressure vessels and heating at 75 °C for 11-14 days gave 9 in yields of 91-

99%.

Porphyrin 9 marks the intersection of the synthetic shown in Scheme 2.5 and the originally

proposed pathway in Scheme 2.2. In an effort to completely circumvent the previous catastrophic

deprotection reaction, a palladium catalyzed coupling of carboxylic acid-bearing 9 and rhodamine

11 was carried out in toluene. Following overnight reflux, TLC analysis of the reaction mixture did

not show obvious formation of the desired product, 13. However, the mass spectrum of this

mixture revealed trace levels of a species with m/z = 1853 matching the molecular weight of 13.

Notwithstanding this positive result, the mixture was predominantly composed of unreacted

porphyrin 9 and dehalogenated rhodamine 11. Shortly after reaching reflux temperature, it was

noted that a suspension of fine particulate appeared to have formed. Given the results of the

reaction, this fine particulate was suspected to be an insoluble cesium carboxylate salt of 9. In

efforts to improve the solubility the reaction was attempted in coordinating organic solvents, THF

and 1,4-dioxane. Surprisingly, no traces of 13 were detected following these reactions. These

results suggested the necessity of a protecting group to eliminate formation of the carboxylate

salt. Synthesis of the photonically controlled antenna-reaction center complex reviewed in

Chapter 1 involved esterification of a porphyrin intermediate with 4-methoxy-α-methylbenzyl

28 28

alcohol, which was later removed by treatment with trifluoroacetic acid in the presence of a huge

excess of a cation scavenger. Coupling of carbodiimide-activated 9 with this alcohol proceeded

smoothly to give 32. As was observed during chromatographic purification of boronate ester 28,

benzyl ester 32 was converted back to the carboxylic acid, 9, if the material was not eluted

quickly enough. Porphyrin 32 was coupled with rhodamine 11 using established Buchwald-

Hartwig aryl amination conditions to give dyad 33.33 Treatment of 33 with trifluoroacetic acid in a

50 mg/mL solution of 1,3,5-trimethoxybenzene in dichloromethane cleanly afforded the free base

dyad 13 bearing a carboxylic acid moiety. From dyad 13, triad 15 was obtained via a well-known

amide forming reaction with anilinofullerene 14.32, 39 Trace quantities of methanol from

chromatographic purification of 13 were found to intercept the carbodiimide-activated acid leading

to formation of the corresponding methyl ester as a minor byproduct. Unfortunately, this impurity

was found to be impossible to separate from triad 15 by column chromatography and so

preventing its formation by removal of trace methanol from 13 was pursued. The most effective

method was found to be repeated dissolution in chloroform followed by distillation of the

chloroform-methanol azeotrope. In spite of these efforts, the methyl ester was still formed in

approximately 4% yield. Separation of this impurity could only be achieved by repeated

preparative thin-layer chromatographic purification. Finally, the pure free base triad 15 was

treated with zinc acetate to afford the target compound, triad 1, in essentially quantitative yield.

Diverging from the synthetic pathway to 1, the model compound dyad 2 was synthesized

from the carboxylic acid-bearing hexaphenylbenzene-porphyrin 9 as shown in Scheme 2.6. The

tert-butylanilide 34 was readily obtained by way of carbodiimide-activatation of 9 in the presence

of 4-tert-butylaniline. This was followed by a Buchwald-Hartwig aryl-amination with rhodamine 11

to give the desired model dyad 2 in modest yield. Model dye 3 was generously provided by Yuichi

Terazono.33

29 29

HN

9

N N

NN OH

O

NH

Zn

O

N N

NN HN

O

NH

Zn

O

N NNN HN

O

N

Zn

O

O

ON

N

O

2

34

i

ii O

ON

N

OBr

11

Scheme 2.6 The synthetic pathway followed to obtain 2: i) EDCI, DMAP, CH2Cl2, RT; ii) Pd(OAc)2, Cs2CO3, P(tBu)3, toluene, reflux.

30 30

Chapter 3

Photophysical and Kinetics Analyses

Photophysical characterization of reaction center and antenna models focuses on the

fate of photo-generated excited states of individual chromophores as they return to the ground

state through various decay pathways. In the absence of interchromophore interactions, the

mechanisms of excited state decay can generally be summarized as internal conversion,

intersystem crossing, and fluorescence. Internal conversion (IC) refers to non-radiative vibrational

decay to the ground state accompanied by the release of heat. Intersystem crossing (ISC) is a

spin-forbidden transition signaling conversion between singlet and triplet states. Fluorescence

involves emission of a photon whose wavelength is representative of the energy of the excited

state to ground state transition from which it arises. In covalently linked multi-chromophore

systems, additional singlet excited state decay pathways are activated and generally include

electron and energy transfer processes.

As has been observed and characterized in many porphyrin donor – fullerene acceptor

systems, the subject of this research, triad 1, was designed to undergo photoinduced electron

transfer to generate a long-lived (nanoseconds) porphyrin radical cation - fullerene radical anion

charge separated state, PZn!+-C60

!−, under neutral conditions. The long lifetime arises from the

ability of fullerenes to stabilize a single negative charge by distribution of the electron density over

the entire carbon framework, thus making them attractive electron acceptors for reaction center

models.50, 51 The charge separated state lifetimes are sufficiently long (nano- to microseconds) to

hypothetically allow for electron transfer to a secondary acceptor in analogy to the photosynthetic

electron transport chain. However, in the absence of a secondary acceptor, the charge separated

state recombines to form the neutral ground state, PZn-C60 accompanied by release of heat.

Marcus theory summarizes the significant contributions to the field of electron transfer

kinetics and thermodynamics put forth by Rudolph A. Marcus.34 The Marcus interpretation of

excited state electron transfer for PZn-C60, shown in Figure 3.1, treats the ground, excited, and

31 31

Figure 3.1 A schematic Marcus theory description of photoinduced electron transfer.

32 32

charge separated states as parabolic potential energy wells. Each well is representative of the

combined PZn and C60 free energies along with contributions from solvation. First, absorption of a

photon promotes an electron from the ground state to an excited state; in this case the lowest

energy singlet excited state is shown. This vertical transition implies that no nuclear motion takes

place. Transfer of an electron from 1PZn to C60 to form PZn!+-C60

!− is then thermodynamically

allowed from this energized state. Electron transfer occurs along the horizontal reaction

coordinate and takes the lowest energy path along which the potential energy surfaces of the

initial and final states intersect. The rate of electron transfer (charge separation) can thus be

defined using the following equations:

𝑘!" =  𝑣  𝑒𝑥𝑝(−∆𝐺!"‡ /𝑅𝑇) (3.1)

∆𝐺!"‡ =   !

!(1 + ∆𝐺°/𝜆)! (3.2)

The term v includes the electronic coupling between the donor and acceptor, λ refers to the total

reorganization energy associated with donor-acceptor and solvent nuclear rearrangements,

ΔG‡CS is the activation energy, and ΔG° represents the thermodynamic driving force for the

reaction. Through these equations, changes in electron transfer rates in relation to solvent

polarity can be rationalized. Polar solvents are able to stabilize PZn!+-C60

!− state more so than

non-polar solvents, represented by a negative vertical shift its potential well. Therefore, the

driving force increases and the activation energy decreases resulting in rate enhancement. For

the case shown, it is also worth noting that the activation energy of charge recombination, ΔG‡CR,

is relatively large giving rise to the long lifetime of PZn!+-C60

!−.

As was found in previous studies involving the rhodamine dye moiety of compounds 1, 2

and 3, the ability of its open form to quench the excited states of covalently linked zinc porphyrins

is attributed to rapid energy transfer and subsequent decay to the ground state.33 A theoretical

model for the decay of singlet excited states by through-space singlet-singlet energy transfer was

33 33

developed by Theodor Förster and has since become hugely influential in fields ranging from

chemical physics to biophysical dynamics.35, 36 The Förster resonance energy transfer (FRET)

model describes transfer of excited state energy via donor-acceptor dipole-dipole interactions.

Essentially, the strength of this interaction depends on resonant coupling between the donor and

acceptor transition dipoles. The transition of the donor excited state decay must energetically

overlap with the transition of the acceptor excited state formation to some degree; the better the

overlap, the stronger the interaction. Förster's theory treats this as a Coulombic interaction

between the oscillating transition dipoles of a single donor electron and a single acceptor

electron. Oscillations of the donor electron induce oscillations in the acceptor electron by through-

spce electrostatic interactions allowing for the passage of energy from one electron to the other.

Thus, the relative orientations of the dipoles and the distance between them also influence the

strength of the interaction. This model allows for prediction of energy transfer rates based on

experimentally measurable properties using the following equation:

𝑘!"#$ =  !"""  !"!"  !!  !!!"#  !!  !!  !!  !!"

!  𝐽 (3.3)

In this equation, κ2 is the dipole-orientation factor, kr is the radiative rate constant of the donor

(calculated from the fluorescence decay lifetime and the fluorescence quantum yield), NA is

Avogadro's number, n is the refractive index of the solvent, RDA is the donor-acceptor center-to-

center separation distance, and J is the overlap integral related to the strength of donor-acceptor

interactions.

Steady-state spectroscopies provide useful information regarding photophysical processes

that can be detected at the equilibrium state of a system. Absorption measurements allow for

observation of transitions from the ground state (S0) to excited states (S1, S2, etc.) induced by

photo-excitation via incremental exposure to a portion of the UV-visible-and near IR light

spectrum. Each absorption band that is recorded, indicating that photons of particular

34 34

wavelengths were absorbed by the sample rather than passing through to the detector, refers to a

particular transition of a particular species within the sample, barring the complete energetic

overlap of transitions from different species. Emission measurements give information regarding

transitions beginning from the first singlet excited state (S1) or triplet (T1) state leading to the

ground state (S0). A sample is continuously excited at a wavelength corresponding to a particular

electronic transition and the photons emitted from the formed excited state are captured by a

photon measuring device. Transitions detected by steady-state methods are often useful as

diagnostic indicators of sample composition and can be used to calculate the relative energy

levels of emissive excited state species.

Time-resolved fluorescence and transient absorption spectroscopies are used to obtain

kinetic information regarding the excited state lifetimes of unique chemical species and the rates

by which they decay. Sufficiently emissive excited states may be monitored by decay of their

fluorescence at a particular wavelength. The decay profiles obtained by this type of measurement

can then be fit by a multi-exponential decay model. Multiple decay components represent

chemically unique species that share a common S1 to S0 transition energy (emission wavelength).

Each component is fit by a decay time constant that represents the S1 lifetime of that species.

The inverse of this lifetime is the sum total of the rate constants of all decay processes available

to that species. In order to detect very short-lived and/or non-emissive transient species and to

deconvolute the decay kinetics of species whose emissions overlap signficantly, transient

absorption spectroscopy is used. In contrast to the aforementioned optical techniques, transient

absorption is readily able to record transitions from S1 to Sn, S1 to T1, T1 to S0, S1 to S0, etc. The

technique used to study triad 1 and model compounds 2 and 3 is known as pump-probe transient

absorption spectroscopy. First, a population of excited states is generated by an intense,

monochromatic 'pump' pulse. Then, following a specified time delay, a broad-spectrum 'probe'

pulse is passed through the sample and the absorption spectrum is captured. Each spectrum

captured is recorded as the difference in absorption collected before and after the pump pulse.

This alternation of pump and probe pulses is repeated to obtain a series of difference absorption

35 35

spectra at varying time delays. For this work, the time delays separating these light pulses were

varied from hundreds of femtoseconds to tens of nanoseconds. Taken together, this series of

difference spectra at varying time delays can be globally fit to produce a set of decay associated

spectra (DAS) that describe the formations and decays of individual transient species including

singlet and triplet excited states and charge separated states. As a whole, the global fitting results

describe photophysical processes such as fluorescence, energy transfer, charge separation,

charge recombination, and intersystem crossing. The information obtained through these

spectroscopic techniques allows for determination of rate constants and mechanisms for the

formations and decays of individual transient chemical species.

Spectroscopic analyses of 1, 2 and 3 were carried out in order to compile kinetic and

mechanistic descriptions of the major photophysical processes that follow photo-excitation of 1

via its primary chromophore, a zinc porphyrin (PZn). The steady-state experiments and the

kinetics interpretation from the total system perspective described in this chapter were

predominantly the work of the author. The time-resolved fluorescence and transient absorption

experiments, data analyses, and mechanistic interpretations were performed by Gerdenis Kodis.

Steady-State Absorption and Emission

The acetic acid (AcOH) mediated conversion of 1c to 1o in dichloromethane (CH2Cl2) was

monitored using steady-state absorption and emission spectroscopies. The absorption spectrum

of 1c shown in Figure 3.2A (solid, black) is composed of maxima typical of a zinc porphyrin (PZn)

at 422 nm (Soret) and 549 and 589 nm (Q-bands). Absorption bands attributed to the closed

rhodamine dye (DC) at 311 and 330 nm and the fullerene (C60) at 706 nm were also observed but

are not shown.33 Addition of AcOH produced a broad absorbance band at ca.

650 nm accompanied by slightly increased absorbance across the visible range. These optical

features are characteristic of the open dye (DO) and indicate formation of 1o.33 Following

acidification, the sample was neutralized with aqueous Na2CO3, filtered through a pad of silica,

36 36

Figure 3.2 Absorption (A) and emission (B) spectra were collected as 1 was titrated with AcOH: 0 M (solid, black), 0.43 M (dash), 0.83 M (dot), 1.2 M (dot-dash), 1.6 M (dot-dash-dot), and 2.3 M (solid, blue). Neutralization with Na2CO3 (aq) resulted in the spectra shown as circles. Spectra have been corrected for volume changes. A The maxima at 422, 559, and 591 nm are characteristic of PZn. Increasing intensity of the feature at 652 nm, diagnostic for DO, signifies formation of 1o. B Emission spectra (λex = 550 nm) show decreasing intensity of 1PZn fluorescence at 598 and 647 nm concomitant with formation of 1o. Maxima at 711 and 793 nm arise from 1C60.

37 37

evaporated to dryness under a stream of argon, and redissolved in CH2Cl2. The resulting

absorption spectrum (open circles) very closely matches the pre-titration spectrum of 1c, thus

indicating reversible interconversion between 1c and 1o. The emission spectrum collected for 1c

(λex = 550 nm) shown in Figure 3.2B (solid, black) features maxima at 605 and 652 nm

characteristic of fluorescence from the zinc porphyrin singlet excited state (1PZn) and at 711 and

793 nm, which are attributed to fluorescence from the fullerene singlet excited state (1C60).

Protonation of the dye with AcOH led to a significant reduction in porphyrin emission intensity

indicating that Do strongly quenches 1PZn. Following neutralization, the emission spectrum

(open circles) closely resembles that of 1c further demonstrating reversible interconversion

between 1c and 1o by protonation/deprotonation of the dye. The slight increase of 1C60 emission

intensity at 700-800 nm relative to that of 1PZn at 600-650 nm can be explained by increased

quenching of 1PZn by charge separation due to rate enhancement in the presence of trace water

or methanol from the work-up procedure.34 This effect is noted in kinetics anlysis discussion vide

infra. However, this may also/instead indicate that minor decomposition occurred during the

experiment.

Time-Resolved Fluorescence

Fluorescence decays (λex = 550 nm, λobs = 650 nm) for 1 and 2 in CH2Cl2 with 0 M, 1.6 M,

and 4.5 M AcOH were recorded using the time correlated single photon counting technique and

the data were fit by multi-exponential decay curves, shown in Figure 3.3, to obtain decay lifetimes

for photophysically unique species (Table 3.1). The decays for 1c and 2c (0 M AcOH) gave

τ1 = 183 ps and 1.97 ns, respectively. The lifetime of 2c is in good agreement with those reported

for similar compounds suggesting that it decays via typical pathways.33 The excited state lifetime

of 1c is an order of magnitude shorter than that of 2c, a result of one or more quenching

processes attributed to interactions between 1PZn and C60. The 183 ps lifetime of 1c is expected

to include the kinetics from charge separation to form the PZn!+-C60

!− charge separated state and

38 38

[AcOH] τ1 (ps) τ2 (ps) χ2

1

0 M 183 (100%) - ~1.1 1.6 M 178 (33%) 17 (67%) ~1.1 4.5 M 160 (15%) 17 (85%) ~1.1

2

0 M 1970 (100%) - ~1.1 1.6 M 1960 (32%) 19 (68%) ~1.1 4.5 M 1970 (13%) 19 (87%) ~1.1

Figure 3.3 Time-resolved fluorescence decays collected at 650 nm for of 1 and 2 in CH2Cl2 (A

and B, respectively) (λex = 420 nm) and their exponential decay fit lines with

[AcOH] = 0 M (circles), 1.6 M (triangles), and 4.5 M (squares).

Table 3.1 Time-resolved fluorescence exponential decay fitting results.

39 39

Figure 3.4 Time-resolved fluorescence decay profile collected at 800 nm 1 in toluene (λex = 550 nm) and the exponential decay fit line. The fit is composed of a rise component with τ = 255 ps, and decay components with τ = 1.34 ns (90%) and 2.18 ns (10%). The rise time is attributed to formation of 1C60 from decay of 1PZn. The 1.34 ns component is characteristic of 1C60 decay and the 2.18 ns component is associated with impurities resulting from sample degradation.

40 40

also, possibly, energy transfer to form 1C60.39, 52-56 The fitting results obtained for 1 and 2 in the

presence of AcOH were each composed of two decay components indicating that the samples

contained mixtures of their closed and open forms, the relative populations of which are reflected

by their associated pre-exponential factors (tabulated as percentages). In each case the longer τ1

lifetimes are attributed to fluorescence decay arising from the closed forms, as they are similar to

the τ1 lifetimes observed in the absence of AcOH. The τ1 lifetimes for 1c are noted to decrease

with increasing AcOH, a result of enhanced rates of charge separation in higher polarity

solutions.34 The shorter components, τ2 = 17 ps and 19 ps, are assigned to decay of 1PZn for 1o

and 2o, respectively. The significant decreases in these lifetimes relative to the τ1 lifetimes are

attributed to the presence of a 1PZn to DO singlet-singlet energy transfer decay pathway for 1o and

2o. These lifetimes are also in good agreement with those previously reported for analogous

compounds exhibiting this specific energy transfer mechanism.33

The fluorescence decay of 1c in toluene was collected to investigate the possibility of

singlet-singlet energy transfer from 1PZn to C60. The decay profile (λex = 550 nm, λobs = 800 nm)

was best fit by a multi-exponential decay, shown in Figure 3.4, giving a τ = 255 ps rise

component and decay components with τ = 1.34 ns (90%) and 2.18 ns (10%). The dominant

1.34 ns component is attributed to decay of 1C60, which coincides with the fluorescence ascribed

to 1C60 in the steady-state emission spectrum of 1c. This lifetime is also in strong agreement with

reported 1C60 lifetimes.52-54, 56 While some portion of 1C60 is likely formed by direct excitation of C60

at 550 nm, the much greater OD550 of PZn relative to that of C60 suggests that some portion of the

1C60 population likely evolved as a result of 1PZn decay. The rise component is attributed to such a

process, for which the mechanism is discussed vide infra.52-54 The minor 2.18 ns component is

similar to the decay lifetime of 1PZn for 2c in CH2Cl2 and so is attributed to decay of 'free'-1PZn

produced by sample decomposition through which some C60 was completely cleaved from- or

otherwise rendered an ineffective 1PZn electron and energy acceptor in the degraded compound.

41 41

Transient Absorption

Pump-probe transient absorption spectroscopy was used to identify and characterize the

kinetics of charge separation, singlet-singlet energy transfer, and other excited state decay

processes associated with 1, 2, and 3. Charge separation was induced in 1c by excitation of PZn

(λex = 550 nm) in CH2Cl2 and the difference absorption spectra across several spectral windows

were collected over 3 ns intervals to monitor decay of 1PZn and formation of the resulting charge

separated state. Global fitting of these data produced DAS showing formation of long-lived PZn!+

and C60!− species evolving from decay of 1PZn. The DAS in Figure 3.5A have two components

showing stimulated emission (SE) from 1PZn decay (dash) at 650 nm with τ = 190 ps leading to

induced absorption (IA) from formation of PZn!+ (solid) at 660 nm that did not decay on the

timescale of this experiment.39, 53, 54 The DAS in Figure 3.5B show formation of a non-decaying IA

at 1000 nm with τ = 190 ps indicative of formation of a long-lived C60!−.39, 53, 54 The fitted lifetimes

of formation for PZn!+ and C60

!− agree with the fluorescence decay lifetimes determined for 1c. The

kinetics of formation of 1C60 via energy transfer from 1PZn may also be a component of these

lifetimes. A longer timescale experiment was performed with a deoxygenated sample of 1c in

CH2Cl2 (λex = 560 nm) to observe decay of the charge separated state by recombination and to

detect for other long-lived species. The DAS in Figure 3.6 show decay of IA (solid) at 650 nm

signifying decay of PZn!+ with τ = 5.19 ns. The non-decaying IA (dot) at 700 nm is attributed to a

3C60 species produced by ISC from 1C60.54, 57

The mechanism of 1C60 formation for 1c in toluene was elucidated by monitoring the

decay of 1PZn by transient absorption spectroscopy (λex = 590 nm). Global analysis of the data

gave a good fit for the time constants obtained from the decay of 1C60 emission and resulted in

the three components DAS shown in Figure 3.7. The (solid) component with τ = 255 ps shows

decay of SE at 600 and 650 nm arising from decay of 1PZn and essentially simultaneous decay of

ground state bleaching (GSB) at 550 and 600 nm signifying reformation of the ground state, PZn.

42 42

Figure 3.5 Decay associated spectra fit to transient absorption data collected from 1c in CH2Cl2 (λex = 550 nm). A The (dash) component represents decay of 1PZn with τ = 190 p and the (solid) component arises from non-decaying PZn

!+ species. B The (dash) component indicates formation of C60

!− with τ = 190 ps that did not decay (solid) within the timescale of this experiment.

43 43

Figure 3.6 Decay associated spectra fit to transient absorption data collected from 1c in deoxygenated CH2Cl2 (λex = 560 nm). The (solid) component shows decay of PZn

!+ with τ = 5.19 ns and the (dot) component arises from a non-decaying 3C60.

44 44

Figure 3.7 DAS fit to transient absorption data collected from 1c in toluene (λex = 590 nm). The (solid) component shows concerted decay of 1PZn and reformation of its ground state with τ = 255 ps. The (dash) component represents formation of 3C60 with τ = 1.34 ns that did not decay (dot) within the time scale of this experiment.

45 45

This concerted return to the ground state is indicative of an energy transfer mechanism, as

formation of a transient charge separated state would show a delay in the decay of GSB.54 The

(dash) component represents decay of 1C60 and formation of 3C60 with a lifetime of 1.34 ns, which

generated a non-decaying IA (dot) from the long-lived 3C60.57

A CH2Cl2/AcOH solution comprised predominantly of 3o was examined using transient

absorption spectroscopy to determine the mechanism by which the singlet excited state of the

open dye (1DO) decays. As no emission could be detected for this species by steady-state or

time-resolved fluorescence measurements, it was expected to be an ultrafast process. The

resulting DAS (λex = 640 nm) shown in Figure 3.8 describe a three-component decay mechanism

fit by time constants τ = 1.6 ps (solid), 2.9 ps (dash), and 12 ps (dot). The 1.6 ps component

(solid) shows decay of SE at 700-770 nm signifying decay of 1DO and formation of GSB at

640 nm attributed to formation of a hot ground state, hotDO, with τ = 1.6 ps.58 This ultrafast

relaxation of 1DO likely occurs via intramolecular internal charge transfer (ICT).58, 59 The other

components show decay of GSB at 630 and 620 nm with τ = 2.9 ps and 12 ps, respectively.

These decays are attributed to vibrational cooling of hotDO, typically the slowest components of

relaxation, which yield the fully relaxed ground state, DO.58, 59 Alternatively, there could be an

inverted kinetics case where the 2.9 ps component is assigned to relaxation of 1DO and the 1.6 ps

component is assigned to cooling. The 12 ps component (dot) displays a slightly blue shifted

GSB, indicative of a vibrational cooling process involving rearrangment of the solvent, and IA

around 700 nm, characteristic of the non-equilibrated ground state where SE from 1DO decay is

observed.59 The most probable pathway for this relaxation process is illustrated in Figure 3.9.

Transient absorption measurements of 1o and 2o were collected to monitor energy

transfer from 1PZn to DO. The DAS for 2o in CH2Cl2 (λex = 420 nm) shown in Figure 3.10A are

composed of two components that describe decay of 1PZn by energy transfer to DO followed by

ultrafast relaxation of 1Do. The (dash) component shows decay of SE at 610 and 660 nm with

τ = 19 ps signifying decay of 1PZn concomitant with formation of 1DO. This time constant is

consistent with the fluorescence decay lifetimes fit for 1o and 2o. Following energy transfer, the

46 46

Figure 3.8 DAS fit to transient absorption data collected from 3o in CH2Cl2 with an excess of AcOH (λex = 640 nm). The solid component arises from decay of 1DO with τ = 1.6 ps. The remaining components arise from vibrational relaxation and cooling of the hot ground state, DO

hot, with τ = 2.9 ps (dash) and 12 ps (dot).

47 47

Figure 3.9 A schematic representation of the relaxation pathway of 1DO. The shapes and vertical positions of the energy wells, the placements and separations of vibrational energy levels, and the relative horizontal positions of the minima were chosen arbitrarily in order to illustrate the most probable relaxation pathway and its corresponding decay liftime assignments based on the transient absorption results for 3o. The curvatures of the non-emissive transitions (shown in red and green) are meant to convey movement along a second reaction coordinate axis representing solvent reorganization.

48 48

Figure 3.10 A DAS fit to transient absorption data collected from 2o in CH2Cl2 with an excess of AcOH (λex = 420 nm). The (dash) component arises from decay of 1PZn with τ = 19 ps. The (solid) component is attributed to decay of 1DO with τ = 1.6 ps. B DAS fit to transient absorption data shown as (dot) were collected from 1o in CH2Cl2 with an excess of AcOH (λex = 420 nm) and could be fit by the same lifetimes as those for 2o. Three-point smoothing of these data gave the (dash) and (solid) lines.

49 49

formation of IA at 610 and 660 nm for which the global analysis gave a good fit for

τ = 1.6 ps (solid) suggests that this component can be ascribed to relaxation of 1DO as observed

for 3o. This component may also include relaxation from the PZn S2 excited state, which has a

lifetime of around 1.5 ps following excitation at the Soret.60 These fitting results were obtained by

fixing τ = 19 ps in order to obtain an acceptable fit with only the 19 ps and 1.6 ps components.

Under identical experimental conditions, the data obtained from 1o gave DAS with similar

features to those of 2o and were fit reasonably well using the same time constants

(Figure 3.10B). For both sets of data, the 19 ps decay is mixed with the longer multi-component

decay process associated with DO as seen in 3o.

Kinetics Analysis

For the calculations of rate constants that follow, the rates of IC, ISC and emission

processes for 1PZn are expressed as a sum total, k0. The rate constants of the decay processes

associated with 1PZn for 1 are influenced significantly by whether or not energy transfer from 1PZn

to C60 occurs. Formation of the 1C60 species, inferred from the time-resolved fluorescence and

transient absorption experiments described, could likely occur by one of two mechanisms other

than direct excitation of C60. The photophysical analysis reported for a structurally analogous

PZn-C60 dyad suggests that the mechanism could involve transient formation of a PZn!+-C60

!−

charge separated state, whose energy level in toluene lies between that of 1PZn and 1C60, followed

by rapid recombination to form PZn-1C60.54 However, the DAS for 1c in toluene does not show

evidence supporting the involvement of such an intermediate. Formation of a charge separated

state would be characterized by formation of IA at around 660 nm from PZn!+ as seen in the

spectra collected in for 1c in CH2Cl2. There is, however, the possibility that the charge separated

state is formed but decays too rapidly to produce any diagnostic transient signals above the

instrument noise level.

The 255 ps time constant, τRise, taken from fluorescence decay and transient absorption

50 50

measurements for 1c in toluene is composed of decay kinetics for 1PZn. The 2.18 ns 1PZn lifetime

found in the fluorescence decay fitting for 1c in toluene was used to determine k0. The rate of

energy transfer to C60 was determined to be k1 = 3.5 x 109 s-1 using Eq 3.4.

𝑘! =  !

!!"#$  −  𝑘! (3.4)

The rate of this proposed energy transfer process was also calculated using the Förster rate

equation, Eq 3.3. The fluorescence quantum yield of 1PZn was taken as 0.04.61 The extinction

coefficient of the functionalized C60 was measured as 2315 M-1 cm-1 for the anilinofullerene

compound mentioned in the synthesis section. The structure of 1c was minimized using the PM6

semiempirical method and the PZn-C60 center-to-center distance of 19 Å was taken as the

interchromophore distance. Assuming a dipole-orientation factor of 2/3 (random orientation), the

rate was determined to be kFRET = 2.8 x 109 s-1. Comparing this to the experimentally determined

rate in question and taking into consideration the assumption of random dipole orientation and

that solvent was not included in the structure minimization, it is reasonable to assign the observed

rise kinetics to through-space singlet-singlet energy transfer as implicated by the transient

absorption dynamics. The rate of 1C60 decay was determined from it's fluorescence lifetime as

k2 = 7.5 x 108 s-1 which includes ISC to form 3C60 and other typical relaxation processes.

The energy level diagram for 1 shown in Figure 3.11 provides a schematic view of the

decay pathways from 1PZn (2.09 eV) leading to the ground state, D-PZn-C60, for which the

remaining rate calculations are described herein. The charge separated state is estimated to lie

1.4 eV above the ground state based on that of a tetra-aryl zinc porphyrin-fullerene dyad in THF

(1.42 eV).53 The energy level for 1C60 (1.75 eV) is an approximate value used for a range of

organic solvents.54 While 3C60 is not included in this scheme, its energy level lies approximately

0.2-0.3 eV below that of 1C60.54 Based on the lowest energy absorption band of DO (652 nm) and

highest energy stimulated emission peak measured for 1DO (~700 nm), its first singlet excited

51 51

Figure 3.11 Energy level diagram showing the pathways of 1PZn decay for 1 in CH2Cl2.

52 52

state is estimated to lie at 1.84 eV. Relaxation of 1DO is represented by a single, complex

transition to the ground state as rate constants for the individual components of this process

cannot be accurately determined from the experimental data. Kinetic treatment of 1DO relaxation

is discussed vide infra.

The following calculations of rate constants are based off of the fitted fluorescence decay

component lifetimes τ1 and τ2 found in Table 3.1. The lifetimes τ1 = 182, 178, and 160 ps for 1c

include kinetics from charge separation, 1PZn to C60 energy transfer, and intrachromophore

relaxation of 1PZn. The 1.97 ns lifetime of 2c in CH2Cl2 was used to determine k0 for 1 in CH2Cl2.

Thus, the rates of charge separation for 1 in CH2Cl2 with 0 M, 1.6 M, and 4.5 M AcOH were

determined to be k3 = 1.5 x 109 s-1, 1.7 x 109 s-1, and 2.3 x 109 s-1, respectively, using Eq 3.5.

𝑘! =  !!!  − 𝑘! − 𝑘! (3.5)

The rate of charge recombination was determined to be k4 = 1.9 x 108 s-1 from the PZn!+-C60

!−

lifetime measured by transient absorption. The shorter fluorescence decay components observed

for 2 are significantly influenced by the rate of 1PZn to DO energy transfer, kEnT2, occurring in the

protonated species 2o. Thus, the rate of energy transfer k = 5.2 x 1010 s-1 in both 1.6 M and

4.5 M AcOH/CH2Cl2 was determined using Eq 3.6.

𝑘 =   !!!  −  𝑘! (3.6)

Similarly, the photophysical dynamics of 1o are heavily influenced by rapid 1PZn to DO energy

transfer as well as kinetics from 1P to C60 energy and electron transfer. The rate constant for

energy transfer to DO for 1o was determined to be k5 = 5.3 x 1010 s-1 in both 1.6 M and

4.5 M AcOH/CH2Cl2 using Eq 3.7.

53 53

𝑘! =  !!!  − 𝑘! − 𝑘! − 𝑘! (3.7)

The transient absorption data for 3o show a three component relaxation mechanism for 1DO of

which the first step is ultrafast decay to a hot ground state with τ = 1.6 ps. While it is likely that

this lifetime includes kinetic contributions from one or both of the subsequent cooling steps, the

rate constant for 1DO relaxation to the ground state can be approximated as the inverse of this

lifetime, which gives k6 = 6.3 x 1011 s-1.

For 1c, the quantum yields of 1PZn to C60 energy transfer (EnT1) and charge separation

(CS) were determined using Eq 3.8 and 3.9 as follows: ΦEnT1 = 0.64 and ΦCS = 0.27 (0 M AcOH),

ΦEnT1 = 0.61 and ΦCS = 0.30 (1.6 M AcOH), and ΦEnT1 = 0.55 and ΦCS = 0.36 (4.5 M AcOH).

Φ!"#$ =  !!

!!!!!!!! (3.8)

Φ!" =  !!

!!!!!!!! (3.9)

For 1o, the quantum yields of 1PZn to C60 energy transfer (EnT1), charge separation (CS) and 1PZn

to Do energy transfer (EnT2) were determined for 1o using Eq 3.10, 3.11, and 3.12 as follows:

ΦEnT1 = 0.060, ΦCS = 0.029, ΦEnT2 = 0.90 (1.6 M AcOH) and ΦEnT1 = 0.059, ΦCS = 0.039, and

ΦEnT2 = 0.89 (4.5 M AcOH).

Φ!"#$ =  !!

!!!!!!!!!!! (3.10)

Φ!" =  !!

!!!!!!!!!!! (3.11)

Φ!"#$ =  !!

!!!!!!!!!!! (3.12)

For both open and closed forms, the yields of charge separation increase with increasing [AcOH]

due to faster rates of charge separation (k3) in more polar solutions. Relative to the yields of 1PZn

54 54

to C60 energy transfer and charge separation for 1c, those of 1o are observed to decrease by one

order of magnitude due to 1PZn to Do energy transfer accounting for roughly 90% of the decay

kinetics.

In addition to the single-molecule kinetic descriptions above, these results can be viewed

from the perspective of the net photophysical behavior of each sample by considering the total

yields of PZn!+-C60

!− and 1C60 versus those of 1DO with respect to total 1PZn (1c + 1o) formed. In the

absence of AcOH this is directly apparent from the quantum yields, which attribute decay of 1PZn

to 64% energy transfer to C60 and 27% charge separation. In the presence of AcOH, the relative

populations of 1c and 1o are reflected by the fluorescence decay component compositions

shown in Table 1. With 1.6 M AcOH the sample was composed of 33% 1c and 67% 1o, which

when multiplied by the quantum yields of their respective decay processes, gives the decay of

total 1PZn by 60% energy transfer to DO, 24% energy transfer to C60, and 12% charge separation.

Similarly, with 4.5 M AcOH the sample was composed of 15% 1c and 85% 1o for which the total

1PZn decays by 76% energy transfer to DO, 13% energy transfer to C60, and 9% charge

separation. Thus, the formation of charge separated states from total 1PZn is reduced by 56% in

1.6 M AcOH and by 67% in 4.5 M AcOH. Additionally, formation of 1C60 from total 1PZn is reduced

by 63% in 1.6 M AcOH and by 80% in 4.5 M AcOH. The yield reductions for charge separation

are lower than those for energy transfer because charge separation becomes more efficient with

higher [AcOH] while the efficiency of energy transfer is assumed to be unaffected. Thus, the

maximum degree of downregulation of charge separation is limited by the [AcOH] required to

push the equilibrium further toward 1o and the rate and relative efficiency of charge separation

that result from the accompanying change in polarity. In a more practical sense, this is further

limited by the maximum [AcOH] that 1 remains stable in; at some point irreversible loss of zinc

from the porphyrin may occur, which would permanently compromise the total active population

of 1.

55 55

Concluding Remarks

Activation of NPQ in the photosynthetic apparatus of green plants has the effect of

decreasing the emission intensities and fluorescence decay lifetimes of light harvesting antenna

complexes, thereby inhibiting their ability to transfer excitation energy to photosynthetic reaction

centers. This regulatory mechanism is activated in proportion to decreases in thylakoid lumen pH

produced by increased photon flux. The efficiency of charge separation with respect to photons

absorbed by the antennae is consequently reduced since much of the absorbed energy is

diverted away from the reaction centers. Deactivation of NPQ follows increase of lumen pH at

low-light levels as the surplus of absorbed energy is consumed. This research demonstrates

analogous responsive, self-regulating reaction center functionality in 1. The spectral analyses

give evidence for i) reversible conversion between unquenched and quenched states that is

dependent on acid concentration, ii) the capacity to form a long-lived charge separated state

under neutral conditions, and iii) significant reduction of charge separation efficiency under acidic

conditions.

Steady-state absorption and emission spectroscopy were used to monitor acidification

and subsequent neutralization of a solution containing 1 to demonstrate reversible quenching of

the zinc porphyrin excited state and dissipation of said energy by a vibrational process. These

responsive and reversible changes indicate that the equilibrium between 1c and 1o is directly

affected by acetic acid concentration. The significant reduction of porphyrin fluorescence

concomitant with formation of 1o is associated with energy transfer from the zinc porphyrin

excited state to the open rhodamine dye via a Förster-type mechanism. This excitation energy is

then rapidly dissipated as heat by a multi-step process. Transient absorption results indicate that

the relaxation mechanism begins with ultrafast decay to a hot ground state followed by rapid

cooling to its true ground state. Kinetics analysis of 1 indicates that the open dye effectively

decreases the yield of charge separation by one order of magnitude on a per-molecule basis, and

was shown to affect the overall population yield of charge separated states relative to zinc

56 56

porphyrin excited states formed by as much as 67%. Furthermore, the total yield of 1C60 was

observed to be reduced by as much as 80% in the presence of AcOH, which would therefore

reduce formation of 3C60 by intersystem crossing. Analogous to the photoproprotective

functionality observed in NPQ, this may serve to prevent degradation of 1 by 1O2. However,

further experimentation is required to confirm that 1 exhibits photoprotective behavior under

acidic conditions.

Together, these results demonstrate regulation of charge separation efficiency using acid

concentration as a control signal, thus providing a functional analog to NPQ. The next logical step

toward introducing additional biomimetic functionality would be to use a light-responsive proton

source rather than titration with acetic acid so that the regulatory behavior could be controlled by

photon flux. Reversible organic photoacids, typically phenolic compounds that behave as weak

acids in the ground state and strong or super acids in the excited state, may be useful for

achieving this goal. However, the rates of proton dissociation and recombination for such

photoacids are generally characterized in aqueous solutions and have been found, by the author,

to slow dramatically in all organic solvents and aqueous mixtures thereof. To circumvent this

issue, forming water-soluble micelles containing triad 1 may allow for the desired proton transfer

dynamics between 1 and a suitable photoacid to be achieved. Usage of micelles would also fill

the auxiliary role of modeling the heterogeneous membrane environment in which photosynthesis

occurs. Thus, the effects of surface charge, detergent rigidity, and micelle composition on the

photophysical dynamics of 1 could also be investigated in such systems.

57 57

References

1. Stephanopoulos, N.; Solis, E.O.; and Stephanopoulos, G. Nanoscale Process Systems Engineering: Toward Molecular Factories, Synthetic Cells, and Adaptive Devices. AIChE Journal, 2005, 51, 1858-1869.

2. Gust, D.; Moore, T.A.; and Moore, A.L. Solar Fuels via Artificial Photosynthesis. Accounts of Chemical Research, 2009, 42, 1890-1898.

3. Gust, D.; Moore, T.A.; and Moore, A.L. Towards Molecular Logic and Artificial Photosynthesis. From non-covalent assemblies to molecular machines, 2010, 321-353.

4. Gust, D.; Moore, T.A.; and Moore, A.L. Mimicking Photosynthetic Solar Energy Transduction. Accounts of Chemical Research, 2001, 34, 40-48.

5. Moore, T.A.; Moore, A.L.; and Gust, D. The design and synthesis of artificial photosynthetic antennas, reaction centres, and membranes. Philosophical Transactions of the Royal Society B, 2002, 357, 1481-1498.

6. Ruban, A.V.; Johnson, M.P.; and Duffy, C.D.P. The photoprotective molecular switch in the photosystem II antenna. Biochimica et Biophysica Acta, 2012, 1817, 167-181.

7. Horton, P.; Ruban, A.V.; and Walters, R.G. Regulation of light harvesting in green plants. Plant Physiology, 1994, 106, 415-420.

8. Telfer, A.; He, W.; and Barber, J. Spectral resolution of more than one chlorophyll electron donor in the isolated Photosystem II reaction center complex. Biochimica et Biophysica Acta, 1990, 1017, 143-151.

9. Barber, J. Molecular Basis of the Vulnerability of Photosystem II to Damage by Light. Australian Journal of Plant Physiology, 1994, 22, 201-208.

10. Nagy, L.; Balint, E.; Barber, J.; Ringler, A.; Cook, K.M.; and Maroti, P. Photoinhibition and Law of Reciprocity in Photosynthetic Reactions of Synechocystic sp. PCC 6803*. Journal of Plant Physiology, 1995, 145, 410-415.

11. Briantais, J.M.; Vernotte, C.; Picaud, M.; and Krause, G.H. Biochimica et Biophysica Acta, 1979, 548, 128-138.

58 58

12. Johnson, M.P.; and Ruban, A.V. Photoprotective energy dissipation in higher plants involves alteration of the excited state energy of the emitting chlorophyll(s) in the light harvesting antenna II (LHCII). The Journal of Biological Chemistry, 2009, 284, 23592-23601.

13. Millineaux, C.W.; Pascal, A.A.; Horton, P.; and Holzwarth, A.R. Excitation-energy quenching in aggregates of the LHC II chlorophyll-protein complex: a time-resolved fluorescence study. 1993,

14. van Oort, B.; van Hoek, A.; Ruban, A.V.; and van Amerongen, H. Aggregation of light-harvesting complex II leads to formation of efficient excitation energy traps in monomeric and trimeric complexes. FEBS Letters, 2007, 581, 3528-3532.

15. Millineaux, C.W.; Ruban, A.V.; and Horton, P. Prompt heat release associated with ΔpH-dependent quenching in spinach thylakoid membranes. Biochimica et biophysica acta. Bioenergetics, 1994, 1185, 119-123.

16. Rees, D.; and Horton, P. The mechanisms of changes in photosystem II efficiency in spinach thylakoids. Biochimica et Biophysica Acta, 1990, 1016, 219-227.

17. Wraight, C.A.; and Crofts, A.R. Energy-Dependent Quenching of Chlorophyll a Fluorescence in Isolated Chloroplasts. European Journal of Biochemistry, 1970, 17, 319-327.

18. Horton, P.; Wentworth, M.; and Ruban, A. Control of the light harvesting function of chloroplast membranes: The LHCII-aggregation model for non-photochemical quenching. FEBS Letters, 2005, 4201-4206.

19. Holzwarth, A.R.; Miloslavina, Y.; Nilkens, M.; and Jahns, P. Identification of two quenching sites active in the regulation of photosynthetic light-harvesting studied by time-resolved fluorescence. Chemical Physics Letters, 2009, 483, 262-267.

20. Rees, D.; Young, A.; Noctor, G.D.; Britton, G.; and Horton, P. Enhancement of the deltapH-dependent dissipation of excitation energy in spinach chloroplasts by light-activation; correlation with the synthesis of zeaxanthin. FEBS Letters, 1989, 256, 85-90.

21. Noctor, G.D.; Rees, D.; Young, A.; and Horton, P. The relationship between zeaxanthin, energy-dependent quenching of chlorophyll fluorescence and the transthylakoid pH-gradient in isolated chloroplasts. Biochimica et Biophysica Acta, 1991, 1057, 320-330.

59 59

22. Zia, A.; Johnson, M.P.; and Ruban, A. Acclimation- and mutation-induced enhancement of PsbS levels affects the kinetics of non-photochemical quenching in Arabidopsis thaliana. Planta, 2011, 233, 1253-1264.

23. Miloslavina, Y.; Wehner, A.; Lambrev, P.; Wientjes, E.; Reus, M.; Garab, G.; Croce, R.; and Holzwarth, A.R. Far-red fluorescence: A direct spectroscopic marker for LHCII oligomer formation in non-photochemical quenching. FEBS Letters, 2008, 582, 3625-3631.

24. Ma, Y.-Z.; Holt, N.E.; Li, X.; Niyogi, K.K.; and Fleming, G.R. Evidence for direct carotenoid involvement in the regulation of photosynthetic light harvesting. Proceedings of the National Academy of Sciences, 2003, 100, 4377-4382.

25. Liao, P.-N.; Holleboom, C.-P.; Wilk, L.; Kuehlbrandt, W.; and Walla, P.J. Correlation of Car S1 -> Chl with Chl -> Car S1 Energy Transfer Supports the Excitonic Model in Quenched Light Harvesting Complex II. Journal of Physical Chemistry B, 2010, 114,

26. Dreuw, A.; Fleming, G.R.; and Gordon-Head, M. Role of electron-transfer quenching of chlorophyll fluorescence by carotenoids in non-photochemical quenching of green plants. Biochemical Society Transactions, 2005, 33, 858-862.

27. Holt, N.E.; Zigmantas, D.; Valkunas, L.; Li, X.; Niyogi, K.K.; and Fleming, G.R. Carotenoid Cation Formation and the Regulation of Photosynthetic Light Harvesting. Science, 2005, 307, 433-436.

28. Ruban, A.V.; Berera, R.; Ilioaia, C.; van Stokkum, I.H.M.; Kennis, J.T.M.; Pascal, A.A.; van Amerongen, H.; Robert, B.; Horton, P.; and van Grongelle, R. Identification of a mechanism of photoprotective energy dissipation in higher plants. Nature, 2007, 450, 575-579.

29. Muller, M.G.; Lambrev, P.; Reus, M.; Wientjes, E.; Croce, R.; and Holzwarth, A.R. Singlet Energy Dissipation in the Photosystem II Light-Harvesting Complex Does Not Involve Energy Transfer to Carotenoids. ChemPhysChem, 2010, 11, 1289-1296.

30. Berera, R.; Herrero, C.; van Stokkum, I.H.M.; Vengris, M.; Kodis, G.; Palacios, R.E.; van Amerongen, H.; van Grondelle, R.; Gust, D.; Moore, T.A.; Moore, A.L.; and Kennis, J.T.M. A simple artificial light-harvesting dyad as a model for excess energy dissipation in oxygenic photosynthesis. Proceedings of the National Academy of Sciences, 2006, 103, 5343-5348.

31. Kodis, G.; Herrero, C.; Palacios, R.; Marino-Ochoa, E.; Gould, S.; de la Garza, L.; van Grondelle, R.; Gust, D.; Moore, T.A.; Moore, A.L.; and Kennis, J.T.M. Light Harvesting

60 60

and Photoprotective Functions of Carotenoids in Compact Artifical Photosynthetic Antenna Designs. Journal of Physical Chemistry B, 2004, 108, 414-425.

32. Straight, S.D.; Kodis, G.; Terazono, Y.; Hambourger, M.; Moore, T.A.; Moore, A.L.; and Gust, D. Self-regulation of photoinduced electron transfer by a molecule nonlinear transducer. Nature Nanotechnology, 2008, 3, 280-283.

33. Terazono, Y.; Kodis, G.; Bhushan, K.; Zaks, J.; Madden, C.; Moore, A.L.; Moore, T.A.; Fleming, G.R.; and Gust, D. Mimicking the Role of the Antenna in Photosynthetic Photoprotection. Journal of the American Chemical Society, 2011, 133, 2916-2922.

34. Marcus, R.A. On the Theory of Oxidation-Reduction Reactions Involving Electron Transfer. I. The Journal of Chemical Physics, 1956, 24, 966-978.

35. Forster, T. Energiewanderung und Fluoreszenz. Naturwissenschaften, 1946, 33, 166-175.

36. Forster, T. Zwischenmolekulare Energiewanderung und Fluoreszenz. Annalen der Physik, 1948, 437, 55-75.

37. Gust, D.; Moore, T.A.; and Moore, A.L. Molecular swithces controlled by light. Chemical Communications, 2006, 1169-1178.

38. Woodroofe, C.C.; Lim, M.H.; Bu, W.; and Lippard, S.J. Synthesis of isomerically pure carboxylae- and sulfonate-substituted xanthene fluorophores. Tetrahedron, 2005, 61, 3097-3105.

39. Terazono, Y.; Liddell, P.A.; Garg, V.; Kodis, G.; Brune, A.; Hambourger, M.; Moore, A.L.; Moore, T.A.; and Gust, D. Artificial photosynthetic antenna-reaction center complexes based on a hexaphenylbenzene core. Journal of Porphyrins and Phthalocyanines, 2005, 09, 706-723.

40. Littler, B.J.; Ciringh, Y.; and Lindsey, J.S. Investigation of Conditions Giving Minimal Scrambling in the Synthesis of trans-Porphyrins from Dipyrromethanes and Aldehydes. Journal of Organic Chemistry, 1999, 64, 2864-2872.

41. Geier, G.R.; and Lindsey, J.S. Effects of aldehyde or dipyrromethane substituents on the reaction course leading to meso-substituted porphyrins. Tetrahedron, 2004, 60, 11435-11444.

61 61

42. Liddell, P.A.; Gervaldo, M.; Bridgewater, J.W.; Keirstead, A.E.; Lin, S.; Moore, T.A.; Moore, A.L.; and Gust, D. Porphyrin-Based Hole Conducting Electropolymer. Chemistry of Materials, 2008, 20, 135-142.

43. Gervaldo, M.; Liddell, P.A.; Kodis, G.; Brennan, B.J.; Johnson, C.R.; Bridgewater, J.W.; Moore, A.L.; Moore, T.A.; and Gust, D. A photo- and electrochemically-active porphyrin-fullerene dyad electropolymer. Photochemical & Photobiological Sciences, 2010, 9, 890-900.

44. Schmitz, R.A.; Liddell, P.A.; Kodis, G.; Kenney, M.J.; Brennan, B.J.; Oster, N.V.; Moore, T.A.; Moore, A.L.; and Gust, D. Synthesis and spectroscopic properties of a soluble semiconducting porphyrin electropolymer. Physical Chemistry Chemical Physics, 2014, 16, 17569-17579.

45. Ingold, K.U. Inhibition of the Autoxidation of Organic Substances in the Liquid Phase. Chemical Reviews, 1961, 61, 563-589.

46. Ingold, K.U.; and Pratt, D.A. Advances in Radical-Trapping Antioxidant Chemistry in the 21st Century: A Kinetics and Mechanisms Perspective. Chemical Reviews, 2014, 114, 9022-9046.

47. Wuts, P.G.M.; and Greene, T.W., Protection for the Amino Group in Greene's Protective Groups in Organic Synthesis. Fourth Edition ed. 2006, Hoboken, NJ, USA: John Wiley & Sons, Inc.

48. Battacharyya, S.; Kibel, A.; Kodis, G.; Liddell, P.A.; Gervaldo, M.; Gust, D.; and Lindsay, S. Optical Modulation of Molecular Conductance. Nano Letters, 2011, 11, 2709-2714.

49. Suzuki, A. Recent advances in the cross-coupling reactions of organoboron derivatives with organic electrophiles, 1995-1998. Journal of Organometallic Chemistry, 1999, 576, 147-168.

50. Imahori, H.; and Sakata, Y. Donor-Linked Fullerenes: Photoinduced Electron Transder and Its Potential Application. Advanced Materials, 1997, 9, 537-546.

51. Imahori, H.; and Sakata, Y. Fullerens as Novel Acceptors in Photosynthetic Electron Transfer. European Journal of Organic Chemisry, 1999, 2445-2457.

52. Kuciauskas, D.; Lin, S.; Seely, G.R.; Moore, A.L.; Moore, T.A.; Gust, D.; Drovetskaya, T.; Reed, C.A.; and Boyd, P.D.W. Energy and Photoinduced Electron Transfer in Porphyrin-Fullerene Dyads. Journal of Physical Chemistry, 1996, 100, 15926-15932.

62 62

53. Imahori, H.; Hagiwara, K.; Aoki, M.; Akiyama, T.; Taniguchi, S.; Okada, T.; Shirakawa, M.; and Sakata, Y. Linkage and Solvent Dependence of Photoinduced Electron Transfer in Zincporphyrin-C60 Dyads. Journal of the American Chemical Society, 1996, 118, 11771-11782.

54. Imahori, H.; El-Khouly, M.E.; Fujitsuka, M.; Ito, O.; Sakata, Y.; and Fukuzumi, S. Solvent Dependence of Charge Separation and Charge Recombination Rates in Porphyrin-Fullerene Dyad. Journal of Physical Chemistry A, 2001, 105, 325-332.

55. Liddell, P.A.; Kodis, G.; Kuciauskas, D.; Andreasson, J.; Moore, A.L.; Moore, T.A.; and Gust, D. Photoinduced electron transfer in a symmetrical diporphyrin-fullerene triad. Physical Chemistry Chemical Physics, 2004, 6, 5509-5515.

56. Prato, M.; and Maggini, M. Fulleropyrrolidines:  A Family of Full-Fledged Fullerene Derivatives. Accounts of Chemical Research, 1998, 31, 519-526.

57. Campidelli, S.; Vasquez, E.; Milic, D.; Prato, M.; Barbera, J.; Guldi, D.M.; Maracaccio, M.; Paolucci, D.; Paolucci, F.; and Deschenaux, R. Liquid-crystalline fullerene-ferrocene dyads. Journal of Materials Chemistry, 2004, 14, 1266-1272.

58. Kovalenko, S.A.; Schanz, R.; Farztinov, V.M.; Hennig, H.; and Ernsting, N.P. Femtosecond relaxation of photoexcited para-nitroaniline: solvation, charge transfer, internal conversion and cooling. Chemical Physics Letters, 2000, 323, 312-322.

59. Martin, M.M.; Plaza, P.; and Meyer, Y.H. Ultrafast intramolecular charge transfer in the merocyanine dye DCM. Chemical Physics, 1995, 192, 367-377.

60. Yu, H.-Z.; Baskin, S.; and Zewail, A.H. Ultrafast Dynamics of Porphyrins in the Condensed Phase: II. Zinc Tetraphenylporphyrin. Journal of Physical Chemistry A, 2002, 106, 9845-9854.

61. Hsiao, J.-S.; Krueger, B.P.; Wagner, R.W.; Johnson, T.E.; Delaney, J.K.; Mauzerall, D.C.; Fleming, G.R.; Lindsey, J.S.; Bocian, D.F.; and Donohoe, R.J. Soluble Synthetic Multiporphyrin Arrays. 2. Photodynamics of Energy-Transfer Processes. Journal of the American Chemical Society, 1996, 118, 11181-11193.

62. Rohand, T.; Dolusic, E.; Ngo, T.H.; Maes, W.; and Dehaen, W. Efficient synthesis of aryldipyrromethanes in water and their application in the synthesis of corroles and dipyrromethenes. ARKIVOC, 2007, 307-324.

63 63

APPENDIX A

SYNTHETIC METHODS AND CHARACTERIZATION DATA

64 64

Materials

All commercially available reagents including 1-(4-methoxyphenyl)ethanol,

1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDCI), 1,3,5-trimethoxybenzene,

2,4,6-trimethylbenzyl alcohol, 4-bromo-iodobenzene, 4-dimethylaminopyridine (DMAP),

4-tert-butylaniline, acetyl chloride, bis(pinacolato)diboron, boron trifluoride diethyl etherate

(BF3•OEt2), cesium carbonate (Cs2CO3), copper(I) iodide (CuI), methyl 4-formylbenzoate,

potassium acetate (KOAc), potassium hydroxide (KOH), pyridine,

tetraphenylcyclopentadieneone, triethylamine (Et3N), trifluoroacetic acid (TFA), trifluoroacetic

anhydride, tripotassium phosphate (K3PO4), tri-tert-butylphosphine (PtBu3), and

zinc acetate dihydrate [Zn(OAc)2•2H2O] were used as received. Palladium catalysts

[1,1’-bis(diphenylphosphino)ferrocene]palladium(II) dichloride dichloromethane complex

[Pd(dppf)Cl2•CH2Cl2], bis(triphenylphosphine)palladium(II) dichloride [PdCl2(PPh3)2],

palladium(II) acetate [Pd(OAc)2], and tetrakis(triphenylphospine)palladium(0) [Pd(PPh3)4]

were purchased from Strem Chemicals, Inc. Dichloromethane (CH2Cl2) was distilled from calcium

hydride when used as a reaction solvent or to prepare optical spectroscopy samples.

Tetrahydrofuran (THF) and toluene were distilled from sodium/benzophenone for use in

reactions. 1,4-Dioxane was distilled from lithium aluminum hydride. Other solvents including

acetone, chloroform (CHCl3), ethanol (EtOH), diethyl ether (Et2O), diphenyl ether (Ph2O), ethyl

acetate (EtOAc), hexanes, and methanol (MeOH) were used as recieved. Column

chromatography was performed using Silicycle SiliaFlash® F60 40-63 µm (230-400 mesh). Thin

layer chromatography was performed using Analtech Silica Gel GHL/GHLF 250 µm plates.

65 65

Analytical Instrumentation and Methods

Nuclear magnetic resonance (NMR) spectra were collected using a Varian MR 400 MHz

spectrometer equipped with a 5 mm liquids H-X broadband probe or a Varian VNMRS 500 MHz

spectrometer equipped with a 5 mm liquids triple resonance 1H-13C-15N probe. Samples were

prepared in deuterochloroform (CDCl3) (Cambridge Isotope Laboratories, Inc.) containing

tetramethylsilane (TMS) (0.03% v/v) as an internal reference. Data was processed using

ACD/NMR Processor Academic Edition. Chemical shifts were referenced to TMS. Mass spectra

were obtained using an Applied Biosystems Voyager-DE STR matrix-assisted laser

desorption/ionization time-of-flight mass spectrometer (MALDI-TOF-MS).

66 66

15-(4-(2-(4-bromophenyl)ethynyl)-phenyl)-5-(4-carbomethoxyphenyl)-10,20-bis(2,4,6-

trimethylbenzene)porphyrin 5. Condensations of 5-mesityldipyrromethane62, 4-(4-

bromophenylethynyl)benzaldehyde 4, and methyl 4-formylbenzoate were carried out following

general literature procedures.40, 41 One method employed BF3•OEt2/EtOH as the acid catalyst in

CHCl3 and gave a 21% yield of 5 in one instance.39 Alternatively, the reaction was catalyzed by

TFA in CH2Cl2. However no final yields were calculated following this method, as the products

were not of adequate purity to carry forward. 1H NMR (400 MHz, CDCl3, δ): -2.64 (br s, 2H, NH),

1.84 (s, 12H, CH3), 2.63 (s, 6H, CH3), 4.11 (s, 3H, OCH3), 7.29 (s, 4H, ArH), 7.56 (m, 4H, ArH),

7.92 (d, J = 8 Hz, 2H, ArH), 8.21 (d, J = 8 Hz, 2H, ArH), 8.32 (d, J = 8.4 Hz, 2H, ArH),

8.43 (d, J = 8.4 Hz, 2H, ArH), 8.72 (m, 6H, βH), 8.80 (d, J = 5 Hz, 2H, βH). MALDI-TOF-MS m/z:

calcd for C60H47BrN4O2 936.29, obsd 936.60.

NH N

HNN O

OBr

67 67

Porphyrin 6. In a flask equipped with a magnetic stir bar, 5 (470 mg, 0.5 mmol) and

tetraphenylcyclopentadieneone (3.5 g, 5 mmol) were dissolved in diphenyl ether (60 mL). The

reaction was refluxed under Ar for 4 h. The solution was concentrated in vacuo and the residue

was chromatographed (flash column, 3:2 CH2Cl2/hexanes). Recrystallization from CH2Cl2/MeOH

gave 523 mg of 6 (81% yield). 1H NMR (400 MHz, CDCl3, δ): -2.74 (br s, 2H, NH),

1.80 (s, 12H, CH3), 2.62 (s, 3H, CH3), 2.63 (s, 3H, CH3), 4.08 (s, 3H, OCH3),

6.90-6.95 (m, 15H, ArH), 7.00 (d, J = 8 Hz, 2H, ArH), 7.08-7.16 (m, 7H, ArH),

7.26-7.29 (m, 6H, ArH), 7.72 (d, J = 8 Hz, 2H, ArH), 8.27 (d, J = 8 Hz, 2H, ArH),

8.33 (d, J = 5 Hz, 1H, βH), 8.41 (m, 3H, ArH and βH), 8.61 (d, J = 5 Hz, 1H, βH),

8.65-8.70 (m, 5H, βH). MALDI-TOF-MS m/z: calcd for C88H67BrN4O2 1293.45, obsd 1292.94.

NH N

HNN O

O

Br

68 68

Porphyrin 7. In a flask equipped with a magnetic stir bar, 6 (400 mg, 0.31 mmol) was

dissolved in CHCl3 (200 mL). To the solution was added Zn(OAc)2·2H2O/MeOH (680 mg/25 mL)

and the reaction was stirred under Ar overnight at room temperature. The solution was washed

with water (x 2) and brine and then concentrated in vacuo. The residue was chromatographed

(flash column, 2:1 CH2Cl2/hexanes to CH2Cl2) to obtain 418 mg of 7 (99% yield). 1H NMR

(400 MHz, CDCl3, δ): 1.81 (s, 12H, CH3), 2.63 (s, 3H, CH3), 2.64 (s, 3H, CH3),

4.09 (s, 3H, OCH3), 6.92-6.97 (m, 15H, ArH), 7.02 (d, J = 8 Hz, 2H, ArH), 7.12-7.17 (m, 7H, ArH),

7.28-7.31 (m, 6H, ArH), 7.74 (d, J = 8 Hz, 2H, ArH), 8.30 (d, J = 8 Hz, 2H, ArH),

8.40 (d, J = 8 Hz, 2H, ArH), 8.45 (d, J = 4 Hz, 1H, βH), 8.53 (d, J = 5 Hz, 1H, βH)

8.71 (d, J = 5 Hz, 1H, βH), 8.75 (d, J = 4 Hz, 2H, βH), 8.78-8.80 (m, 3H, βH). MALDI-TOF-MS

m/z: calcd for C88H65BrN4O2Zn 1354.36, obsd 1354.98.

N N

NN O

O

Br

Zn

69 69

Porphyrin 8. In a flask equipped with a magnetic stir bar, 7 (175 mg, 0.129 mmol),

Pd(OAc)2 (1.5 mg, 0.006 mmol), and Cs2CO3 (420 mg, 1.29 mmol), were dissolved in Ar bubbled

toluene (20 mL). To the solution was added PtBu3 (12 uL, 1M in toluene). The reaction was

refluxed under Ar overnight. The cooled reaction mixture was filtered through a short plug of silica

gel and concentrated in vacuo. The residue was chromatographed (flash column,

9:1 CH2Cl2/hexanes to CH2Cl2) to obtain 157 mg of 8 (87% yield). 1H NMR (400 MHz, CDCl3, δ):

1.79 (s, 6H, CH3), 1.81 (s, 6H, CH3), 2.63 (s, 3H, CH3), 2.64 (s, 3H, CH3), 3.76 (s, 3H, OCH3),

4.09 (s, 3H, OCH3), 5.45 (br s, 1H, NH), 6.72 (d, J = 8 Hz, 2H, ArH), 6.75 (d, J = 9 Hz, 2H, ArH),

6.89-7.04 (m, 19H, ArH), 7.14 (m, 5H, ArH), 7.20 (d, J = 8 Hz, 2H, ArH), 7.27 (s, 2H, ArH),

7.29 (s, 2H, ArH), 7.74 (d, J = 8 Hz, 2H, ArH), 8.30 (d, J = 7.9 Hz, 2H, ArH),

8.40 (d, J = 7.9 Hz, 2H, ArH), 8.52 (d, J = 4 Hz, 1H, βH), 8.59 (d, J = 4 Hz, 1H, βH),

8.68 (d, J = 5 Hz, 1H, βH), 8.72 (d, J = 5 Hz, 1H, βH), 8.76 (d, J = 4 Hz, 2H, βH),

8.79 (d, J = 5 Hz, 2H, βH). MALDI-TOF-MS m/z: calcd for C95H73N5O3Zn 1396.50, obsd 1398.08.

N N

NN O

O

NH

Zn

O

70 70

Porphyrin 9. In a flask equipped with a magnetic stir bar, 8 (128 mg, 0.092 mmol) was

dissolved in THF/MeOH (2/1, 180 mL). The solution was bubbled with Ar at 0 °C for 15 min

followed by addition of aqueous KOH (10% w/v, 30 mL). The reaction was stirred under Ar

overnight at room temperature. The solution was washed with water (2 x) and aqueous citric

acid (0.1 M), and then concentrated in vacuo. The residue was chromatographed (flash column,

8% MeOH/CH2Cl2) to obtain 115 mg of 9 (91% yield). 1H NMR (400 MHz, CDCl3, δ):

1.81 (s, 6H, CH3), 1.82 (s, 6H, CH3), 2.64 (s, 3H, CH3), 2.65 (s, 3H, CH3), 3.77 (s, 3H, OCH3),

6.73 (d, J = 8 Hz, 2H, ArH), 6.76 (d, J = 9 Hz, 2H, ArH), 6.88-7.05 (m, 19H, ArH),

7.14 (m, 5H, ArH), 7.21 (d, J = 8 Hz, 2H, ArH), 7.28 (s, 2H, ArH), 7.30 (s, 2H, ArH),

7.75 (d, J = 8 Hz, 2H, ArH), 8.36 (d, J = 7.9 Hz, 2H, ArH), 8.50 (d, J = 7.9 Hz, 2H, ArH),

8.53 (d, J = 5 Hz, 1H, βH), 8.60 (d, J = 4 Hz, 1H, βH), 8.69 (d, J = 4.4 Hz, 1H, βH),

8.73 (d, J = 4.4 Hz, 1H, βH), 8.78 (d, J = 5 Hz, 2H, βH), 8.82 (d, J = 4 Hz, 2H, βH).

MALDI-TOF-MS m/z: calcd for C94H71N5O3Zn 1382.49, obsd 1383.93.

N N

NN OH

O

NH

Zn

O

71 71

Porphyrin 10. In a flask equipped with a magnetic stir bar, 9 (115 mg, 0.083 mmol), 2,4,6-

trimethylbenzyl alcohol (14 mg, 0.116 mmol), DMAP (20 mg, 0.166 mmol), and EDCI (32 mg,

0.166 mmol) were dissolved in CH2Cl2 (18 mL). The reaction was cooled to 0 °C and bubbled

with Ar for 15 min. Stirring was continued for 10 min at 0 °C before the reaction was brought to

room temperature and stirred for an additional 24 h. The solution was diluted with CH2Cl2,

washed with water, and concentrated in vacuo. The residue was chromatographed (flash column,

3:1 CH2Cl2/hexanes to CH2Cl2) to obtain 87 mg of 10 (69% yield). 1H NMR (400 MHz, CDCl3, δ):

1.79 (s, 6H, CH3), 1.80 (s, 6H, CH3), 2.32 (s, 3H, CH3), 2.54 (s, 6H, CH3), 2.63 (s, 3H, CH3),

2.64 (s, 3H, CH3), 3.74 (s, 3H, OCH3), 5.43 (br s, 1H, NH), 5.55 (s, 2H, CH2),

6.71 (d, J = 8 Hz, 2H, ArH), 6.74 (d, J = 9 Hz, 2H, ArH), 6.88-7.03 (m, 21H, ArH),

7.14 (m, 5H, ArH), 7.20 (d, J = 8 Hz, 2H, ArH), 7.27 (s, 2H, ArH), 7.28 (s, 2H, ArH),

7.74 (d, J = 8 Hz, 2H, ArH), 8.25 (d, J = 8 Hz, 2H, ArH), 8.35 (d, J = 8 Hz, 2H, ArH),

8.52 (d, J = 4.9 Hz, 1H, βH), 8.59 (d, J = 4.4 Hz, 1H, βH), 8.68 (d, J = 4.4 Hz, 1H, βH),

8.72 (d, J = 4.9 Hz, 1H, βH), 8.75 (d, J = 4.4 Hz, 2H, βH), 8.77 (d, J = 4.4 Hz, 2H, βH).

N N

NN O

O

NH

Zn

O

72 72

Dyad 12. In a flask equipped with a magnetic stir bar, 10 (133 mg. 0.088 mmol), rhodamine

11 (72 mg, 0.117 mmol), Pd(OAc)2 (15 mg, 0.066 mmol), and Cs2CO3 (238 mg, 0.730 mmol),

were dissolved in toluene (10 mL). The solution was cooled to 0 °C and bubbled with Ar for

15 min. To the solution was added PtBu3 (135 uL, 1M in toluene). The reaction was refluxed

under Ar overnight. The cooled reaction mixture was diluted with CH2Cl2, washed water, and then

concentrated in vacuo. The residue was chromatographed twice (flash column, 3% MeOH/CH2Cl2

then 6% acetone/CH2Cl2) to obtain 99 mg of 12 (55% yield). 1H NMR (400 MHz, CDCl3, δ):

1.69 (s, 6H, CH3), 1.80 (s, 6H, CH3), 2.33 (s, 3H, CH3), 2.48 (s, 3H, CH3), 2.55 (s, 6H, CH3), 2.64

(s, 3H, CH3), 3.09 (t, J = 9 Hz, 4H, CH2), 3.50 (s, 3H, OCH3), 3.92 (m, 4H, CH2), 5.58 (s, 2H,

CH2), 6.60 (d, J = 9 Hz, 2H, ArH), 6.74-7.23 (m, 48H, ArH), 7.28 (s, 2H, ArH),

7.50 (d, J = 2 Hz, 1H, ArH), 7.76 (d, J = 8 Hz, 2H, ArH), 8.24 (d, J = 8.4 Hz, 2H, ArH),

8.36 (d, J = 8.4 Hz, 2H, ArH), 8.53 (d, J = 5 Hz, 1H, βH), 8.65 (d, J = 5 Hz, 1H, βH),

8.68-8.77 (m, 6H, βH). MALDI-TOF-MS m/z: calcd for C140H107N7O6Zn 2046.76, obsd 2048.21.

N NNN O

O

N

Zn

O

O

ON

N

O

73 73

Dyad 13. In a flask equipped with a magnetic stir bar, 12 (70 mg, 0.034 mmol) was

dissolved in 10 mL of an Ar bubbled 1:1 mixture of TFA and CH2Cl2. The reaction was stirred

under argon for 1.5 h. The reaction mixture was diluted to 100 mL with CH2Cl2, washed with

water, saturated aqueous NaHCO3 (x 2), aqueous citric acid (0.5 M), and brine, and then

concentrated in vacuo. The residue was chromatographed multiple times to give 53 mg of a

mixture containing 13. MALDI-TOF-MS m/z: calcd for C130H97N7O6 1852.75, obsd 1853.13

(100%), 1985.27 (86%), 2118.40 (23%).

NH NHNN OH

O

N O

O

ON

N

O

74 74

4-(4-formylphenylethynyl)-N-(4-methoxyphenyl)aniline 17. In a flask equipped with a

magnetic stir bar, 4-ethynyl-N-(4-methoxyphenyl)aniline 1633 (1.00 g, 4.50 mmol),

4-iodobenzaldehyde (1.07 g, 4.60 mmol), and CuI (48 mg, 0.23 mmol) were dissolved in

THF/Et3N (12 mL/1.5 mL). The solution was bubbled with Ar at 0 °C for 30 min, PdCl2(PPh3)2

(80 mg, 0.11 mmol) was added, and bubbling continued for 15 min. The reaction was stirred

overnight under Ar at room temperature. The solution was diluted to 100 mL with CH2Cl2, filtered

through a pad of Celite rinsing with CH2Cl2, and concentrated in vacuo. The residue was

chromatographed twice (flash column, 7:3 then 9:1 CH2Cl2/hexanes) and recrystallized from

CH2Cl2/hexanes to obtain 345 mg of 17 (23% yield). 1H NMR (400 MHz, CDCl3, δ): 3.82 (s, 3H,

OCH3), 5.70 (br s, 1H, NH), 6.83 (d, J = 8.6 Hz, 2H, ArH), 6.90 (d, J = 9 Hz, 2H, ArH),

7.12 (d, J = 8.6 Hz, 2H, ArH), 7.39 (d, J = 9 Hz, 4H, ArH), 7.63 (d, J = 8 Hz, 2H, ArH),

7.84 (d, J = 8 Hz, 2H, ArH), 10.00 (s, 1H, CH).

HN

O

H

O

75 75

N-(4-(4-formylphenyl)ethynylphenyl)-N-(4-methoxyphenyl)trifluoroacetamide 18. In a flask

equipped with a magnetic stir bar, 17 (100 mg, 0.31 mmol) and pyridine (98 µL, 1.22 mmol) were

dissolved in CH2Cl2 (31 mL) and placed under Ar. Trifluoroacetic anhydride (420 µL,

1M in CH2Cl2) was added at 0 °C and the reaction was brought to room temperature. Additional

pyridine (98 µL, 1.22 mmol) and trifluoroacetic anhydride/CH2Cl2 (420 µL) were added after 2 h

and stirring continued for 10 min. The solution was diluted to 50 mL with CH2Cl2, washed with

saturated aqueous NaHCO3 (2 x), and then concentrated in vacuo. The residue was and

chromatographed (flash column, 1:1 Et2O/hexanes) to obtain 120 mg of 18 (93% yield). 1H NMR

(400 MHz, CDCl3, δ): 3.83 (s, 3H, OCH3), 6.94 (d, J = 9 Hz, 2H, ArH), 7.24 (d, J = 8 Hz, 2H, ArH),

7.30 (d, J = 8 Hz, 4H, ArH), 7.56 (d, J = 8 Hz, 2H, ArH), 7.66 (d, J = 8.2 Hz, 2H, ArH),

7.87 (d, J = 8.2 Hz, 2H, ArH), 10.02 (s, 1H, CH).

N

O

H

O

F3CO

76 76

4-(4-bromophenylethynyl)-N-(4-methoxyphenyl)aniline 23. In a flask equipped with a

magnetic stir bar, 4-ethynyl-N-(4-methoxyphenyl)aniline 1633 (446 mg, 2.00 mmol),

4-bromo-iodobenzene (566 mg, 2.00 mmol), and CuI (19 mg, 0.10 mmol) were dissolved in

THF/Et3N (9.5 mL/0.5 mL). The solution was bubbled with Ar at 0 °C for 15 min, PdCl2(PPh3)2

(35 mg, 0.050 mmol) was added, and bubbling continued for 15 min. The reaction was stirred

overnight under Ar at room temperature. The solution was diluted with CH2Cl2 (50 mL), filtered

through a pad of Celite rinsing with CH2Cl2, and then concentrated in vacuo. The residue was

chromatographed (flash column, 2:1 CH2Cl2/hexanes) and recrystallized from CH2Cl2/hexanes to

obtain 475 mg of 23 (63% yield). 1H NMR (400 MHz, CDCl3, δ): 3.81 (s, 3H, OCH3),

5.64 (br s, 1H, NH), 6.82 (d, J = 9 Hz, 2H, ArH), 6.89 (d, J = 8.6 Hz, 2H, ArH),

7.10 (d, J = 9 Hz, 2H, ArH), 7.35 (dd, J = 8.6, 2.3 Hz, 4H, ArH), 7.45 (d, J = 8.6 Hz, 2H, ArH).

HN

O

Br

77 77

N-(4-(4-bromophenylethynyl)phenyl)-N-(4-methoxyphenyl)trifluoroacetamide 24. In a

flask equipped with a magnetic stir bar, 23 (40 mg, 0.11 mmol) and pyridine (68 µL, 0.85 mmol)

were dissolved in CH2Cl2 (20 mL) and placed under Ar. Trifluoroacetic anhydride (420 µL,

1M in CH2Cl2) was added at 0 °C and the reaction was stirred for 10 min before being brought to

room temperature. The solution was concentrated to 10 mL in vacuo and chromatographed

(gravity column, CH2Cl2) to obtain 50 mg of 24 (quantitative). 1H NMR (400 MHz, CDCl3, δ):

3.83 (s, 3H, OCH3), 6.92 (d, J = 9 Hz, 2H, ArH), 7.22-7.28 (m, 4H, ArH),

7.37 (d, J = 8 Hz, 4H, ArH), 7.47-7.53 (m, 4H, ArH).

N

O

BrF3C

O

78 78

N-(4''-bromo-3',4',5',6'-tetraphenyl-[1,1':2',1''-terphenyl]-4-yl)-N-(4-

methoxyphenyl)trifluoroacetamide 25. In a flask equipped with a magnetic stir bar, 24 (20 mg,

0.042 mmol) and tetraphenylcyclopentadieneone (81 mg, 0.21 mmol) were dissolved in Ph2O

(2 mL). The reaction was refluxed under Ar for 6 h. Additional tetraphenylcyclopentadieneone

(78 mg, 0.20 mmol) was added and reflux continued for 3 h. The reaction mixture was

concentrated in vacuo and the residue was chromatographed (flash column, CH2Cl2/hexanes: 3/2

to 9/1) to isolate a single band that yielded 26 mg of material. The mass spectrum indicated that

25 was present in the material however, the 1H NMR was nondescript indicating the presence of

substantial impurities.

N

O

Br

O

CF3

79 79

N-(4-(4-bromophenylethynyl)phenyl)-N-(4-methoxyphenyl)acetamide 26. In a flask

equipped with a magnetic stir bar, 23 (333 mg, 0.881 mmol) and pyridine (0.71 mL, 8.8 mmol)

were dissolved in CH2Cl2 (65 mL) and placed under Ar. Acetyl chloride (125 µL, 1.75 mmol) was

added in two equal portions 15 min apart at 0 °C. The reaction was stirred for 45 min. The

solution was washed with saturated aqueous NaHCO3 (x 2), dried over Na2SO4, and then

concentrated in vacuo. The residue was chromatographed (flash column, 8% EtOAc/CH2Cl2) to

obtain 368 mg of 26 (99% yield). 1H NMR (400 MHz, CDCl3, δ): 2.06 (s, 3H, CH3),

3.82 (s, 3H, OCH3), 6.92 (d, J = 9 Hz, 2H, ArH), 7.18 (d, J = 9 Hz, 2H, ArH),

7.24 (d, J = 8.5 Hz, 2H, ArH), 7.36 (d, J = 8.5 Hz, 2H, ArH), 7.47 (d, J = 8.5 Hz, 4H, ArH).

N

O

Br

O

80 80

N-(4''-bromo-3',4',5',6'-tetraphenyl-[1,1':2',1''-terphenyl]-4-yl)-N-(4-

methoxyphenyl)acetamide 27. In a flask equipped with a magnetic stir bar and an air condenser,

26 (355 mg, 0.845 mmol) and tetraphenylcyclopentadieneone (3.25 g, 8.46 mmol) were dissolved

in Ph2O (40 mL) and the reaction was refluxed under Ar for 19 h. The solution was concentrated

in vacuo and the residue was chromatographed (flash column, 1:4 EtOAc/toluene). The material

was recrystallized from refluxing toluene (x 2) to obtain 494 mg of 27 as a 1:1 complex with

toluene (67% yield). 1H NMR (400 MHz, CDCl3, δ): 1.82 (br s, 3H, CH3),

2.35 (s, 3H, toluene-CH3), 3.78 (s, 3H, OCH3), 6.68 (d, J = 7.3 Hz, 2H, ArH),

6.76-6.98 (m, 32H, ArH), 7.14-7.18 (m, 3H, toluene-ArH), 7.24-7.27 (m, 2H, toluene-ArH).

MALDI-TOF-MS m/z: calcd for C51H38BrNO2 775.21, obsd 777.48.

N

O

O

Br

81 81

Boronate Ester 28. In a heavy-walled tube equipped with a magnetic stir bar, 27 (389 mg,

0.448 mmol), bis(pinacolato)diboron (152 mg, 0.600 mmol) and KOAc (245 mg, 2.50 mmol) were

dissolved in 1,4-dioxane (10 mL). The solution was bubbled with Ar at 0 °C for 10 min, then

Pd(dppf)Cl2•CH2Cl2 (18 mg, 0.025 mmol) was added, and bubbling continued for 10 min. The

tube was sealed with a PTFE screw plug and the reaction was held at 100 °C for 36 h. The

solution was suspended in water, extracted with CH2Cl2 (x 3), and the combined extracts were

washed with brine and then concentrated in vacuo. The residue was chromatographed

(flash column, 1:3 EtOAc/toluene) to obtain 261 mg of 28 (71% yield). 1H NMR

(400 MHz, CDCl3, δ): 1.27 (s, 12H, CH3), 1.76 (br s, 3H, CH3), 3.76 (s, 3H, OCH3),

6.70-6.95 (m, 32H, ArH). MALDI-TOF-MS m/z: calcd for C57H50BNO4 823.38, obsd 823.62.

N

O

O

BOO

82 82

Porphyrin 31. In a heavy-walled tube equipped with a magnetic stir bar, 28 (124 mg,

0.15 mmol), (5-bromo-15-(4-carbomethoxyphenyl)-10,20-bis(2,4,6-

trimethylphenyl)porphyrinato)zinc(II)48 30 (124 mg, 0.150 mmol), and K3PO4 (636 mg, 3.00 mmol)

were dissolved in toluene (30 mL). The solution was bubbled with Ar at 0 °C for 20 min, then

Pd(PPh3)4 (17 mg, 0.015 mmol) was added, and bubbling was continued for 15 min. The tube

was sealed with a PTFE screw plug and the reaction was refluxed for 3 d. The solution was

filtered through Celite rinsing with 10% methanol/CH2Cl2 and then concentrated in vacuo. The

residue was chromatographed (flash column, 5% EtOAc/CH2Cl2) to obtain 160 mg of 31

(74% yield). 1H NMR (400 MHz, CDCl3, δ): 1.80 (s, 6H, CH3), 1.81 (s, 6H, CH3),

1.92 (s, 3H, OCH3), 2.61 (br s, 3H, CH3), 2.65 (s, 3H, CH3), 3.23 (br s, 3H, CH3),

4.09 (s, 3H, OCH3), 6.42-7.29 (m, 36H, ArH), 7.71 (d, J = 7 Hz, 2H, ArH),

8.3 (d, J = 8 Hz, 2H, ArH), 8.39-8.44 (m, 3H, βH and ArH), 8.59 (d, J = 4 Hz, 1H, βH),

8.72 (d, J = 4 Hz, 2H, βH), 8.75-8.80 (m, 4H, βH). MALDI-TOF-MS m/z:

calcd for C97H75N5O4Zn 1437.41, obsd 1439.81. UV-Vis (CH2Cl2): λmax, nm 422, 549, 589.

N N

NN O

O

N

Zn

O

O

83 83

Porphyrin 9. In a heavy walled pressure vessel equipped with a magnetic stir bar, 31

(75 mg, 0.052 mmol) was dissolved in THF (5 mL) followed by addition of KOH/MeOH

(2 M, 15 mL). The solution was bubbled with Ar at 0 °C for 30 min, brought to room temperature,

and the vessel was flame-sealed under low vacuum. The reaction was heated at 75 °C for 14 d.

The solution was diluted with CH2Cl2 (50 mL) and washed with brine. The aqueous layer was

extracted with CH2Cl2 (50 mL) and the combined organic layers were washed with aqueous

citric acid (0.2 M, 150 mL) then concentrated in vacuo. The residue was loaded onto a pad of

silica gel, flushed with (x 5) with CH2Cl2, and eluted with 10% MeOH/CH2Cl2 to obtain 71 mg of 9

(99% yield). 1H NMR (400 MHz, CDCl3, δ): 1.81 (s, 6H, CH3), 1.82 (s, 6H, CH3),

2.64 (s, 3H, CH3), 2.65 (s, 3H, CH3), 3.76 (s, 3H, OCH3), 6.71 (d, J = 9 Hz, 2H, ArH),

6.78 (d, J = 9 Hz, 2H, ArH), 6.87-7.04 (m, 19H, ArH), 7.14 (m, 5H, ArH),

7.21 (d, J = 8 Hz, 2H, ArH), 7.28 (s, 2H, ArH), 7.30 (s, 2H, ArH), 7.75 (d, J = 8 Hz, 2H, ArH),

8.36 (d, J = 8.2 Hz, 2H, ArH), 8.50 (d, J = 8.2 Hz, 2H, ArH), 8.53 (d, J = 4.4 Hz, 1H, βH),

8.60 (d, J = 4.8 Hz, 1H, βH), 8.69 (d, J = 4.4 Hz, 1H, βH), 8.73 (d, J = 4.8 Hz, 1H, βH),

8.78 (d, J = 4.8 Hz, 2H, βH), 8.82 (d, J = 4.8 Hz, 2H, βH). MALDI-TOF-MS m/z:

calcd for C94H71N5O3Zn 1381.48, obsd 1383.74. UV-Vis (CH2Cl2): λmax, nm 422, 549, 589.

N N

NN OH

O

NH

Zn

O

84 84

Porphyrin 32. In a flask equipped with a magnetic stir bar, 9 (63 mg, 0.046 mmol),

1-(4-methoxyphenyl)ethanol (128 µL, 0.909 mmol), DMAP (56 mg, 0.46 mmol), and (41 mg,

0.23 mmol) were dissolved in CH2Cl2 (20 mL). The reaction was stirred overnight at room

temperature. The solution was diluted with CH2Cl2, washed with saturated aqueous NaHCO3, and

then concentrated in vacuo. The residue was chromatographed (x 2) (gravity column,

4:1 CH2Cl2/hexanes) to obtain 40 mg of 32 (58% yield). 1H NMR (400 MHz, CDCl3, δ):

1.8 (m, 15H, CH3), 2.63 (s, 3H, CH3), 2.64 (s, 3H, CH3), 3.76 (s, 3H, OCH3), 3.84 (s, 3H, OCH3),

5.44 (br s, 1H, NH), 6.27 (q, J = 7 Hz, 1H, CH), 6.72 (d, J = 8 Hz, 2H, ArH), 6.76 (d, J = 9 Hz, 2H,

ArH), 6.81-7.04 (m, 21H, ArH), 7.14 (m, 5H, ArH), 7.20 (d, J = 8 Hz, 2H, ArH), 7.27 (s, 2H, ArH),

7.29 (s, 2H, ArH), 7.53 (d, J = 8 Hz, 2H, ArH), 7.74 (d, J = 8 Hz, 2H, ArH), 8.28 (s, J = 8.1 Hz, 2H,

ArH), 8.40 (d, J = 8.1 Hz, 2H, ArH), 8.52 (d, J = 4.8 Hz, 1H, βH), 8.59 (d, J = 4.8 Hz, 1H, βH),

8.68 (d, J = 4.8 Hz, 1H, βH), 8.72 (d, J = 4.8 Hz, 1H, βH), 8.75 (d, J = 4 Hz, 2H, βH),

8.78 (d, J = 4 Hz, 2H, βH). MALDI-TOF-MS m/z: calcd for C103H81N5O4Zn 1515.56, obsd 1517.63.

UV-Vis (CH2Cl2): λmax, nm 422, 549, 588.

N N

NN O

O

NH

Zn

O

O

85 85

Dyad 33. In a flask equipped with a magnetic stir bar, 10 (39 mg, 0.026 mmol),

rhodamine 11 (24 mg, 0.039 mmol), Pd(OAc)2 (2.9 mg, 0.013 mmol), and Cs2CO3

(84 mg, 0.26 mmol) were dissolved in toluene (3 mL). The solution was bubbled with Ar at 0 °C

for 20 min, an Ar balloon was attached, and PtBu3 (26 µL, 1 M in toluene) was added. The

reaction was slowly brought to reflux and stirred overnight. The solution was poured over a pad of

silica gel, eluted with 10% MeOH/CH2Cl2, and concentrated in vacuo. The residue was

chromatographed (flash column, 3-5% acetone/CH2Cl2) to obtain 42 mg of 33 (79% yield).

1H NMR (400 MHz, CDCl3, δ): 1.69 (s, 6H, CH3), 1.80 (m, 9H, CH3), 2.48 (s, 3H, CH3),

2.64 (s, 3H, CH3), 3.09 (t, J = 8 Hz, 4H, CH2), 3.50 (s, 3H, OCH3), 3.84 (s, 3H, OCH3),

3.86-3.96 (m, 4H, CH2), 6.27 (q, J = 7 Hz, 1H, CH), 6.60 (d, J = 9 Hz, 2H, ArH),

6.74-7.21 (m, 48H, ArH), 7.28 (s, 2H, ArH), 7.5 (d, J = 2 Hz, 1H, ArH), 7.53 (d, J = 9 Hz, 2H, ArH),

7.76 (d, J = 8 Hz, 2H, ArH), 8.26 (d, J = 8.4 Hz, 2H, ArH), 8.40 (d, J = 8.4 Hz, 2H, ArH),

8.53 (d, J = 5 Hz, 1H, βH), 8.65 (d, J = 5 Hz, 1H, βH), 8.68-8.79 (comp, 6H, βH).

MALDI-TOF-MS m/z: calcd for C139H105N7O7Zn 2047.74.74, obsd 2050.23.

UV-Vis (CH2Cl2): λmax, nm 336, 422, 549, 588.

N NNN O

O

N

Zn

OO

O

ON

N

O

86 86

Dyad 13. In a flask equipped with a magnetic stir bar, 33 (42 mg, 0.02 mmol) and

1,3,5-trimethoxybenzene (1 g, 6 mmol) were dissolved in CH2Cl2 (10 mL). TFA/CH2Cl2

(2% v/v, 10 mL) was added and the reaction was stirred for 5 h at room temperature. TFA

(200 µL) was added and stirring continued for 5 h. The solution was washed with saturated

aqueous NaHCO3 (x 2) and concentrated in vacuo. The residue was loaded onto a short pad of

silica gel and flushed with CH2Cl2 (1.5 L) then 0.5% MeOH/CH2Cl2 (300 mL) and eluted with

10% MeOH/CH2Cl2 to obtain 34 mg of 13 (89% yield). 1H NMR (400 MHz, CDCl3, δ):

-2.70 (br s, 2H, NH), 1.70 (s, 6H, CH3), 1.82 (s, 6H, CH3), 2.47 (s, 3H, CH3), 2.64 (s, 3H, CH3),

3.10 (t, J = 8 Hz, 4H, CH2), 3.57 (s, 3H, OCH3), 3.86-3.97 (m, 4H, CH2),

6.67 (d, J = 9 Hz, 2H, ArH), 6.75-7.22 (m, 48H, ArH), 7.29 (s, 2H, ArH),

7.54 (d, J = 2 Hz, 1H, ArH), 7.75 (d, J = 8 Hz, 2H, ArH), 8.33 (d, J = 8 Hz, 2H, ArH),

8.45 (d, J = 5 Hz, 1H, βH), 8.51 (d, J = 9 Hz, 2H, ArH), 8.57 (d, J = 4 Hz, 1H, βH),

8.62-8.73 (comp, 6H, βH). MALDI-TOF-MS m/z: calcd for C130H97N7O6 1851.75, obsd 1852.80.

UV-Vis (CH2Cl2): λmax, nm 420, 515, 549, 591, 647.

NH NHNN OH

O

N O

O

ON

N

O

87 87

Triad 15. In a flask equipped with a magnetic stir bar, 13 (34 mg, 0.018 mmol),

anilinofullerene 1439 (20 mg, 0.023 mmol), DMAP (9 mg, 0.07 mmol), and EDCI

(7 mg, 0.04 mmol) were dissolved in CH2Cl2 (4 mL). The reaction was stirred under Ar for 24 h at

room temperature. The solution was diluted with CH2Cl2 (50 mL), washed with aqueous citric acid

(0.5 M) and saturated aqueous NaHCO3, and then concentrated in vacuo. The residue was

chromatographed (thin layer, 5% EtOAc/CH2Cl2) to obtain 20 mg of 14 (40% yield).

1H NMR (400 MHz, CDCl3, δ): -2.71 (br s, 2H, NH), 1.69 (s, 6H, CH3), 1.81 (s, 6H, CH3),

2.46 (s, 3H, CH3), 2.63 (s, 3H, CH3), 2.82 (s, 3H, NCH3), 3.08 (t, J = 9 Hz, 4H, CH2),

3.57 (s, 3H, OCH3), 3.85-3.95 (m, 4H, CH2), 4.23 (d, J = 9 Hz, 1H, CH), 4.93-4.98 (m, 2H, CH2),

6.67 (d, J = 9 Hz, 2H, ArH), 6.74-7.22 (comp, 45H, ArH), 7.28 (s, 2H, ArH),

7.53 (d, J = 2 Hz, 1H, ArH), 7.53 (d, J = 9.3 Hz, 2H, ArH), 7.75 (d, J = 8 Hz, 2H, ArH),

7.88 (br s, 3H, ArH), 8.20-8.23 (m, 3H, ArH and NH), 8.28 (d, J = 8 Hz, 2H, ArH),

8.44 (d, J = 4 Hz, 1H, βH), 8.56 (d, J = 5 Hz, 1H, βH), 8.61-8.70 (m, 6H, βH). MALDI-TOF-MS

m/z: calcd for C199H107N9O5 2701.84, obsd 2704.31. UV-Vis (CH2Cl2): λmax, nm 308, 332, 420,

515, 549, 591, 646.

NH NHNN HN

O

N O

O

ON

N

O

N

88 88

Triad 1. In a flask equipped with a magnetic stir bar, 15 (19 mg, 0.0070 mmol) was

dissolved in CH2Cl2 (4 mL). ZnOAc2•2H2O/MeOH (50 mg/mL, 300 µL) was added and the

reaction was stirred under Ar for 24 h at room temperature. The solution was diluted with CH2Cl2,

washed with water and saturated NaHCO3, and concentrated in vacuo. The residue was loaded

onto a short plug of silica gel, flushed with CH2Cl2 (300 mL), and eluted with 10% MeOH/CH2Cl2

to obtain 20 mg of 1 (quantitative). 1H NMR (400 MHz, CDCl3, δ): 1.67 (s, 6H, CH3),

1.78 (s, 6H, CH3), 2.48 (s, 3H, CH3), 2.61 (s, 3H, CH3), 2.72 (s, 3H, NCH3),

3.05 (t, J = 8 Hz, 4H, CH2), 3.54 (s, 3H, OCH3), 3.82-3.92 (m, 4H, CH2),

4.08 (d, J = 10 Hz, 1H, CH), 4.77-4.83 (m, 2H, CH2), 6.64 (d, J = 9 Hz, 2H, ArH),

6.70-7.25 (m, 50H, ArH), 7.48 (d, J = 2 Hz, 1H, ArH), 7.74-7.83 (m, 4H, ArH),

8.09-8.19 (m, 4H, ArH), 8.37 (br s, 1H, NH), 8.53 (d, J = 5 Hz, 1H, βH),

8.64 (d, J = 5 Hz, 1H, βH), 8.67 (d, J = 4 Hz, 1H, βH), 8.70-8.75 (m, 5H, βH). MALDI-TOF-MS

m/z: calcd for C199H105N9O5Zn 2763.75, obsd 2767.08. UV-Vis (CH2Cl2): λmax, nm 311, 330, 422,

549, 589.

N NNN HN

O

N

Zn

O

O

ON

N

O

N

89 89

Porphyrin 34. In a flask equipped with a magnetic stir bar, 9 (30 mg, 0.022 mmol),

4-tert-butylaniline, DMAP (12 mg, 0.098 mmol), and EDCI (8 mg, 0.044 mmol) were dissolved in

CH2Cl2 (10 mL). The reaction was stirred under Ar for 24 h at room temperature. The solution

was applied directly to a short column of silica gel, eluted with CH2Cl2 to 10% MeOH/CH2Cl2, and

concentrated in vacuo. The residue was chromatographed (gravity column, 1% EtOAc/CH2Cl2) to

obtain 29 mg of 34 (87% yield). 1H NMR (400 MHz, CDCl3, δ): 1.37 (s, 9H, CH3),

1.80 (s, 6H, CH3), 1.81 (s, 6H, CH3), 2.63 (s, 3H, CH3), 2.64 (s, 3H, CH3), 3.76 (s, 3H, OCH3),

5.44 (s, 1H, NH), 6.71 (d, J = 9 Hz, 2H, ArH), 6.76 (d, J = 9 Hz, 2H, ArH),

6.88-7.04 (m, 19H, ArH), 7.14 (m, 5H, ArH), 7.20 (d, J = 8 Hz, 2H, ArH), 7.28 (s, 2H, ArH),

7.29 (s, 2H, ArH), 7.47 (d, J = 9 Hz, 2H, ArH), 7.66 (d, J = 9 Hz, 2H, ArH),

7.75 (d, J = 8 Hz, 2H, ArH), 8.05 (s, 1H, NH), 8.17 (d, J = 8.2 Hz, 2H, ArH),

8.33 (d, J = 8.2 Hz, 2H, ArH), 8.52 (d, J = 4.8 Hz, 1H, βH), 8.59 (d, J = 4.4 Hz, 1H, βH),

8.68 (d, J = 4.4 Hz, 1H, βH), 8.72 (d, J = 4.8 Hz, 1H, βH), 8.77 (d, J = 5 Hz, 2H, βH),

8.81 (d, J = 4 Hz, 2H, βH). MALDI-TOF-MS m/z: calcd for C104H84N6O2Zn 1513.60, obsd 1514.61.

UV-Vis (CH2Cl2): λmax, nm 421, 549, 586.

N N

NN HN

O

NH

Zn

O

90 90

Dyad 2. In pressure tube equipped with a magnetic stir bar, 34 (28 mg, 0.018 mmol),

rhodamine 11 (17 mg, 0.028 mmol), Pd(OAc)2 (2 mg, 0.009 mmol), and Cs2CO3 (59 mg,

0.18 mmol) were dissolved in toluene (2 mL). The solution was bubbled with Ar at 0 °C for

15 min, PtBu3 (18 µL, 1 M in toluene) was added and bubbling continued for an additional 15 min.

The tube was sealed with a PTFE screw plug, heated slowly to 115 °C, and stirred for 36 h. The

solution was applied directly to a pad of silica gel, eluted with 10% MeOH/CH2Cl2, and then

concentrated in vacuo. The residue was chromatographed (gravity column,

5-7.5% EtOAc/CH2Cl2) to obtain 10 mg of 2 (27% yield). 1H NMR (400 MHz, CDCl3, δ):

1.38 (s, 9H, CH3), 1.70 (s, 6H, CH3), 1.81 (s, 6H, CH3), 2.48 (s, 3H, CH3), 2.64 (s, 3H, CH3),

3.08 (t, J = 8 Hz, 4H, CH2), 3.50 (s, 3H, OCH3), 3.86-3.95 (m, 4H, CH2),

6.60 (d, J = 9 Hz, 2H, ArH), 6.73-7.21 (m, 46H, ArH), 7.29 (s, 2H, ArH), 7.47-7.50 (m, 3H, ArH),

7.69 (d, J = 8 Hz, 2H, ArH), 7.76 (d, J = 8 Hz, 2H, ArH), 8.07 (s, 1H, NH),

8.18 (d, J = 7.9 Hz, 2H, ArH), 8.32 (d, J = 7.9 Hz, 2H, ArH), 8.54 (d, J = 5 Hz, 1H, βH),

8.66 (d, J = 4 Hz, 1H, βH), 8.70-8.81 (m, 6 H, βH). MALDI-TOF-MS m/z: calcd for

C140H108N8O5Zn 2045.78, obsd 2046.82. UV-Vis (CH2Cl2): λmax, nm 308, 334, 421, 511, 549.

N NNN HN

O

N

Zn

O

O

ON

N

O

91 91

APPENDIX B

METHODS FOR OPTICAL SPECTROSCOPY

92 92

Optical spectroscopy samples were prepared using CH2Cl2 distilled from CaH2. Sample

concentrations for steady-state absorption, emission and time-resolved fluorescence were ca.

10-6 M. Transient absorption samples were ca. 10-4 M. Glacial acetic acid was used to produce

the open, protonated forms of 1, 2, and 3.

Steady-State Spectroscopy

Absorption spectra were measured on Shimadzu UV2100U UV-vis and UV-3101PC

UV-Vis-NIR spectrometers. Fluorescence spectra were measured using a Photon Technology

International MP-1 spectrometer and corrected for detection system response. Excitation was

provided by a 75 W xenon-arc lamp and single grating monochromator. Fluorescence was

detected 90o to the excitation beam via a single grating monochromator and an R928

photomultiplier tube having S 20 spectral response and operating in the single photon counting

mode.

Time-Resolved Fluorescence

Fluorescence decay measurements were performed by the time-correlated single-photon-

counting method. The excitation source was a fiber supercontinuum laser based on a passive

mode-locked fiber laser and a high-nonlinearity photonic crystal fiber supercontinuum generator

(Fianium SC450). The laser provides 6-ps pulses at a repetition rate variable from 0.1-40 MHz.

The laser output was sent through an Acousto-Optical Tunable Filer (Fianium AOTF) to obtain

excitation pulses at desired wavelength of ca. 450-900 nm. Fluorescence emission was detected

at the magic angle (54.7°) using a double grating monochromator (Jobin Yvon Gemini-180) and a

microchannel plate photomultiplier tube (Hamamatsu R3809U-50). The instrument response

function was 35-55 ps. The spectrometer was controlled by software based on the LabView

93 93

programming language and data acquisition was done using a single photon counting card

(Becker-Hickl, SPC-830).

Transient Absorption

The femtosecond transient absorption apparatus consisted of a kHz pulsed laser source

and a pump-probe optical setup. Laser pulses of 100 fs at 800 nm were generated from an

amplified, mode-locked Titanium Sapphire kHz laser system (Millennia/Tsunami/Spitfire, Spectra

Physics). Part of the laser pulse energy was sent through an optical delay line and focused on a 3

mm sapphire plate to generate a white light continuum for the probe beam. The remainder of the

pulse energy was used to pump an optical parametric amplifier (Spectra Physics) to generate

excitation pulses, which were selected using a mechanical chopper. The white light generated

was compressed by prism pairs (CVI) before passing through the sample. The polarization of the

pump beam was set to the magic angle relative to the probe beam and its intensity adjusted using

a continuously variable neutral density filter. The white light probe was dispersed by a

spectrograph (300 line grating) onto a charge-coupled device (CCD) camera (DU420, Andor

Tech.). The final spectral resolution was about 2.3 nm over a 300 nm spectral region. Instrument

response function was ca. 150 fs. Nanosecond time scale measurements were collected using an

EOS spectrometer from Ultrafast Systems (IRF ~800 ps); excitation was from the same optical

parametric amplifier as descibed above.

Data analysis was carried out using locally written software (ASUFIT) developed in

MATLAB (Mathworks Inc.). Decay-associated spectra were obtained by fitting the transient

absorption or fluorescence change curves over a selected wavelength region simultaneously as

described by the parallel kinetic model:

(1)

94 94

where ΔA(λ,t) is the observed absorption (or fluorescence) change at a given wavelength at time

delay t and n is the number of kinetic components used in the fitting. A plot of Ai(λ) vs.

wavelength is called a decay-associated spectrum (DAS), and represents the amplitude spectrum

of the ith kinetic component, which has a lifetime of τi. Random errors associated with the reported

lifetimes obtained from fluorescence and transient absorption measurements were typically ≤ 5%.